Lidé

doc. Ing. Daniel Novák, Ph.D.

Všechny publikace

Classifying and Scoring Major Depressive Disorders by Residual Neural Networks on Specific Frequencies and Brain Regions

  • Autoři: Kang, C., doc. Ing. Daniel Novák, Ph.D., Yao, X., Xie, J., Hu, Y.
  • Publikace: IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2023, 31 2964-2973. ISSN 1534-4320.
  • Rok: 2023
  • DOI: 10.1109/TNSRE.2023.3293051
  • Odkaz: https://doi.org/10.1109/TNSRE.2023.3293051
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Major Depressive Disorder (MDD) – can be evaluated by advanced neurocomputing and traditional machine learning techniques. This study aims to develop an automatic system based on a Brain-Computer Interface (BCI) to classify and score depressive patients by specific frequency bands and electrodes. In this study, two Residual Neural Networks (ResNets) based on electroencephalogram (EEG) monitoring are presented for classifying depression (classifier) and for scoring depressive severity (regression). Significant frequency bands and specific brain regions are selected to improve the performance of the ResNets. The algorithm, which is estimated by 10-fold cross-validation, attained an average accuracy rate ranging from 0.371 to 0.571 and achieved average Root-Mean-Square Error (RMSE) from 7.25 to 8.41. After using the beta frequency band and 16 specific EEG channels, we obtained the best-classifying accuracy at 0.871 and the smallest RMSE at 2.80. It was discovered that signals extracted from the beta band are more distinctive in depression classification, and these selected channels tend to perform better on scoring depressive severity. Our study also uncovered the different brain architectural connections by relying on phase coherence analysis. Increased delta deactivation accompanied by strong beta activation is the main feature of depression when the depression symptom is becoming more severe. We can therefore conclude that the model developed here is acceptable for classifying depression and for scoring depressive severity. Our model can offer physicians a model that consists of topological dependency, quantified semantic depressive symptoms and clinical features by using EEG signals. These selected brain regions and significant beta frequency bands can improve the performance of the BCI system for detecting depression and scoring depressive severity.

Fuzzy Windows with Gaussian Processed Labels for Ordinal Image Scoring Tasks

  • DOI: 10.3390/app13064019
  • Odkaz: https://doi.org/10.3390/app13064019
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    In this paper, we propose a Fuzzy Window with the Gaussian Processed Label (FW-GPL) method to mitigate the overlap problem in the neighboring ordinal category when scoring images. Many published conventional methods treat this challenge as a traditional regression problem and make a strong assumption that each ordinal category owns an adequate intrinsic rank to outline its distribution. Our FW-GPL method aims to refine the ordinal label pattern by using two novel techniques: (1) assembling fuzzy logic to the fully connected layer of convolution neural networks and (2) transforming the ordinal labels with a Gaussian process. Specifically, it incorporates a heuristic fuzzy logic from the ordinal characteristic and simultaneously plugs in ordinal distribution shapes that penalize the difference between the targeted label and its neighbors to ensure a concentrated regional distribution. Accordingly, the function of these proposed windows is leveraged to minimize the influence of majority classes that mislead the prediction of minority samples. Our model is specifically designed to carefully avoid partially missing continuous facial-age segments. It can perform competitively when using the whole continuous facial-age dataset. Extensive experimental results on three facial-aging datasets and one ambiguous medical dataset demonstrate that our FW-GPL can achieve compelling performance results compared to the State-Of-The-Art (SOTA).

Comparing Reminders Sent via SMS Text Messaging and Email for Improving Adherence to an Electronic Health Program: Randomized Controlled Trial

  • Autoři: Kulhánek, A., Lukavská, K., Gabrhelík, R., doc. Ing. Daniel Novák, Ph.D., Burda, V., Prokop, J.
  • Publikace: JMIR mHealth and uHealth. 2022, 10(3), ISSN 2291-5222.
  • Rok: 2022
  • DOI: 10.2196/31040
  • Odkaz: https://doi.org/10.2196/31040
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Background: eHealth interventions can help people change behavior (eg, quit smoking). Reminders sent via SMS text messaging or email may improve the adherence to web-based programs and increase the probability of successful behavior change; however, it is unclear whether their efficiency is affected by the modality of the communication channel. Objective: A 2-armed randomized control trial was conducted to compare the effect of providing reminders via SMS text messaging versus email on the adherence to an eHealth program for smoking cessation and on the probability to initiate a quit attempt. Methods: Smokers were recruited via an internet-based advertisement. A total of 591 participants who diverted from intended use of the program (ie, failed to log on to a session) were automatically randomized to the experimental (SMS text messaging reminder, n=304) or the active comparator (email reminder, n=287) group. Results: Unexpectedly, we found that the mode of reminder delivery did not significantly affect either the adherence, namely the number of completed program sessions, with the SMS text messaging reminder group showing a mean of 4.30 (SD 3.24) and the email reminder group showing a mean of 4.36 (SD 3.27) (t586=0.197, P=.84, and Cohen d=0.016), or the outcome, namely the quit smoking attempt rate (34.2% in the SMS text messaging group vs 31.7% in the email group; χ21=0.4, P=.52). Secondary analyses showed that age, gender, and education had significant effects on program adherence and education on the outcome. Moreover, we found a significant interaction effect between the mode of reminder delivery and gender on program adherence, suggesting that the effectiveness of SMS text message reminders might be different for females and males. However, this particular finding should be treated with care as it was based on post hoc subgroup analysis. Conclusions: This study indicates that the modality of user reminders to log on increased neither the program adherence nor the probability of quitting smoking. This suggests that program developers may save costs using emails instead of SMS text messaging reminders.

Managing Diabetes Using Mobiab: Long-Term Case Study of the Impact of a Mobile App on Self-management

  • DOI: 10.2196/36675
  • Odkaz: https://doi.org/10.2196/36675
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Background This paper describes the development of a mobile app for diabetes mellitus (DM) control and self-management and presents the results of long-term usage of this system in the Czech Republic. DM is a chronic disease affecting large numbers of people worldwide, and this number is continuously increasing. There is massive potential to increase adherence to self-management of DM with the use of smartphones and digital therapeutics interventions. Objective This study aims to describe the process of development of a mobile app, called Mobiab, for DM management and to investigate how individual features are used and how the whole system benefits its long-term users. Using at least 1 year of daily records from users, we analyzed the impact of the app on self-management of DM. Methods We have developed a mobile app that serves as an alternative form to the classic paper-based protocol or diary. The development was based on cooperation with both clinicians and people with DM. The app consists of independent individual modules. Therefore, the user has the possibility to use only selected features that they find useful. Mobiab was available free of charge on Google Play Store from mid-2014 until 2019. No targeted recruitment was performed to attract users. Results More than 500 users from the Czech Republic downloaded and signed up for the mobile app. Approximately 80% of the users used Mobiab for less than 1 week. The rest of the users used it for a longer time and 8 of the users produced data that were suitable for long-term analysis

Motor activity patterns can distinguish between inter-episode bipolar disorder patients and healthy controls

  • DOI: 10.1017/S1092852920001777
  • Odkaz: https://doi.org/10.1017/S1092852920001777
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    OBJECTIVE: Bipolar disorder (BD) is linked to circadian rhythm disruptions resulting in aberrant motor activity patterns. We aimed to explore whether motor activity alone, as assessed by longitudinal actigraphy, can be used to classify accurately BD patients and healthy controls into their respective groups. METHODS: Ninety-day actigraphy records from 25 inter-episode BD patients (i.e. Montgomery-Asberg Depression Rating Scale (MADRS) and Young Mania Rating Scale (YMRS) <15) and 25 sex- and age-matched healthy controls (HC), were used in order to identify latent actigraphic biomarkers capable of discriminating between BD patients and HC. Mean values and time-variations of a set of standard actigraphy features were analysed and further validated using the random forest classifier. RESULTS: Using all actigraphy features, this method correctly assigned 88% (sensitivity=85%, specificity=91%) of BD patients and HC to their respective group. The classification success may be confounded by differences in employment between BD patients and HC. When motor activity features resistant to the employment status were used (the strongest feature being time variation of intradaily variability, Cohen’s d=1.33), 79% of the subjects (sensitivity=76%, specificity=81%) were correctly classified. CONCLUSION: A machine learning actigraphy-based model was capable of distinguishing between inter-episode BD patients and HC solely on the basis of motor activity. The classification remained valid even when features influenced by employment status were omitted. The findings suggest that temporal variability of actigraphic parameters may provide discriminative power for differentiating between BD patients and HC while being less affected by employment status.

Validity of the Aktibipo Self-rating Questionnaire for the Digital Self-assessment of Mood and Relapse Detection in Patients With Bipolar Disorder: Instrument Validation Study

  • Autoři: Anyz, J., Bakstein, E., Dally, A., Kolenic, M., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: JMIR Mental Health. 2021, 8(8), ISSN 2368-7959.
  • Rok: 2021
  • DOI: 10.2196/26348
  • Odkaz: https://doi.org/10.2196/26348
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Background: Self-reported mood is a valuable clinical data source regarding disease state and course in patients with mood disorders. However, validated, quick, and scalable digital self-report measures that can also detect relapse are still not available for clinical care.

Brain Networks of Maintenance, Inhibition and Disinhibition During Working Memory

  • Autoři: Kang, C., Li, Y., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: IEEE Transactions on Neural Systems and Rehabilitation Engineering. 2020, 2020(7), 1518-1527. ISSN 1534-4320.
  • Rok: 2020
  • DOI: 10.1109/TNSRE.2020.2997827
  • Odkaz: https://doi.org/10.1109/TNSRE.2020.2997827
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    During working memory tasks, the brain of humans will process information relying on some mechanisms, including rehearsal, inhibition, disinhibition, and maintenance. By figure out the functional networks of the cerebral, scientists can develop advanced computing approaches to solve some practical problems.

Classification of Actigraphy Records from Bipolar Disorder Patients Using Slope Entropy: A Feasibility Study

  • DOI: 10.3390/e22111243
  • Odkaz: https://doi.org/10.3390/e22111243
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Bipolar Disorder (BD) is an illness with high prevalence and a huge social and economic impact. It is recurrent, with a long-term evolution in most cases. Early treatment and continuous monitoring have proven to be very effective in mitigating the causes and consequences of BD. However, no tools are currently available for a massive and semi-automatic BD patient monitoring and control. Taking advantage of recent technological developments in the field of wearables, this paper studies the feasibility of a BD episodes classification analysis while using entropy measures, an approach successfully applied in a myriad of other physiological frameworks. This is a very difficult task, since actigraphy records are highly non-stationary and corrupted with artifacts (no activity). The method devised uses a preprocessing stage to extract epochs of activity, and then applies a quantification measure, Slope Entropy, recently proposed, which outperforms the most common entropy measures used in biomedical time series. The results confirm the feasibility of the approach proposed, since the three states that are involved in BD, depression, mania, and remission, can be significantly distinguished.

Image-Based Subthalamic Nucleus Segmentation for Deep Brain Surgery with Electrophysiology Aided Refinement

  • DOI: 10.1007/978-3-030-60946-7_4
  • Odkaz: https://doi.org/10.1007/978-3-030-60946-7_4
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Identification of subcortical structures is an essential step in surgical planning for interventions such as the deep brain stimulation (DBS), in which permanent electrode is implanted in a precisely defined location. For refinement of the target localisation and compensation of brain shift occurring during the surgery, intra-operative electrophysiological recording using microelectrodes is usually undertaken. In this paper, we present a multimodal method that consists of a) subthalamic nucleus (STN) segmentation from magnetic resonance T2 images using 3D active contour fitting and b) a subsequent brain shift compensation step, increasing the accuracy of microelectrode placement localisation by the probabilistic electrophysiology-based fitting. The method is evaluated on a data set of 39 multi-electrode trajectories from 20 patients undergoing DBS surgery for Parkinson’s disease in a leave-one-subject-out scenario. The performance comparison shows increased sensitivity and slightly decreased specificity of STN identification using the individually-segmented 3D contours, compared to electrophysiology-based refinement of a standard 3D atlas. To achieve accurate segmentation from the low-resolution clinical T2 images, a more sophisticated approach, including shape priors and intensity model, needs to be implemented. However, the presented approach is a step towards automatic identification of microelectrode recording sites and possibly also an assistive system for the DBS surgery.

Influence of glucometric ‘dynamical’ variables on duodenal‐jejunal bypass liner (DJBL) anthropometric and metabolic outcomes

  • Autoři: Colás, A., Vavrela, M., Mráz, M., doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Vigil, L., Beneš, M., Pelikánová, T., Haluzík, M., Burda, V., Vargas, B.
  • Publikace: Diabetes/Metabolism Research and Reviews. 2020, 36(4), ISSN 1520-7552.
  • Rok: 2020
  • DOI: 10.1002/dmrr.3287
  • Odkaz: https://doi.org/10.1002/dmrr.3287
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Background The endoscopically implanted duodenal‐jejunal bypass liner (DJBL) is an attractive alternative to bariatric surgery for obese diabetic patients. This article aims to study dynamical aspects of the glycaemic profile that may influence DJBL effects. Methods Thirty patients underwent DJBL implantation and were followed for 10 months. Continuous glucose monitoring (CGM) was performed before implantation and at month 10. Dynamical variables from CGM were measured: coefficient of variation of glycaemia, mean amplitude of glycaemic excursions (MAGE), detrended fluctuation analysis (DFA), % of time with glycaemia under 6.1 mmol/L (TU6.1), area over 7.8 mmol/L (AO7.8) and time in range. We analysed the correlation between changes in both anthropometric (body mass index, BMI and waist circumference) and metabolic (fasting blood glucose, FBG and HbA1c) variables and dynamical CGM‐derived metrics and searched for variables in the basal CGM that could predict successful outcomes. Results There was a poor correlation between anthropometric and metabolic outcomes. There was a strong correlation between anthropometric changes and changes in glycaemic tonic control (∆BMI‐∆TU6.1: rho = − 0.67, P < .01) and between metabolic outcomes and glycaemic phasic control (∆FBG‐∆AO7.8: r = .60, P < .01). Basal AO7.8 was a powerful predictor of successful metabolic outcome (0.85 in patients with AO7.8 above the median vs 0.31 in patients with AO7.8 below the median: Chi‐squared = 5.67, P = .02). Conclusions In our population, anthropometric outcomes of DJBL correlate with improvement in tonic control of glycaemia, while metabolic outcomes correlate preferentially with improvement in phasic control. Assessment of basal phasic control may help in candidate profiling for DJBL implantation.

Automated atlas fitting for deep brain stimulation surgery based on microelectrode neuronal recordings

  • DOI: 10.1007/978-981-10-9023-3_19
  • Odkaz: https://doi.org/10.1007/978-981-10-9023-3_19
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    The deep brain stimulation (DBS) is a treatment technique for late-stage Parkinson’s disease (PD), based on chronic electrical stimulation of neural tissue through implanted electrodes. To achieve high level of symptom suppression with low side effects, precise electrode placement is necessary, although difficult due to small size of the target nucleus and various sources of inaccuracy, especially brain shift and electrode bending. To increase accuracy of electrode placement, electrophysiological recording using several parallel microelectrodes (MER) is used intraoperatively in most centers. Location of the target nucleus is identified from manual expert evaluation of characteristic neuronal activity. Existing studies have presented several models to classify individual recordings or trajectories automatically. In this study, we extend this approach by fitting a 3D anatomical atlas to the recorded electrophysiological activity, thus adding topological information. Methods: We developed a probabilistic model of neuronal activity in the vicinity the subthalamic nucleus (STN), based on normalized signal energy. The model is used to find a maximum-likelihood transformation of an anatomical surface-based atlas to the recorded activity. The resulting atlas fit is compared to atlas position estimated from pre-operative MRI scans. Accuracy of STN classification is then evaluated in a leave-one-subject-out scenario using expert MER annotation. Results: In an evaluation on a set of 27 multi-electrode trajectories from 15 PD patients, the proposed method showed higher accuracy in STN-nonSTN classification (88.1%) compared to the reference methods (78.7%) with an even more pronounced advantage in sensitivity (69.0% vs 44.6%). Conclusion: The proposed method allows electrophysiology-based refinement of atlas position of the STN and represents a promising direction in refining accuracy of MER localization in clinical DBS setting, as well as in research of DBS mechanisms.

Influence of duodenal-jejunal implantation on glucose dynamics: A pilot study using different nonlinear methods

  • DOI: 10.1155/2019/6070518
  • Odkaz: https://doi.org/10.1155/2019/6070518
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Diabetes is a disease of great and rising prevalence, with the obesity epidemic being a significant contributing risk factor. Duodenal–jejunal bypass liner (DJBL) is a reversible implant that mimics the effects of more aggressive surgical procedures, such as gastric bypass, to induce weight loss. We hypothesized that DJBL also influences the glucose dynamics in type II diabetes, based on the induced changes already demonstrated in other physiological characteristics and parameters. In order to assess the validity of this assumption, we conducted a quantitative analysis based on several nonlinear algorithms (Lempel–Ziv Complexity, Sample Entropy, Permutation Entropy, and modified Permutation Entropy), well suited to the characterization of biomedical time series. We applied them to glucose records drawn from two extreme cases available of DJBL implantation: before and after 10 months. The results confirmed the hypothesis and an accuracy of 86.4% was achieved with modified Permutation Entropy. Other metrics also yielded significant classification accuracy results, all above 70%, provided a suitable parameter configuration was chosen. With the Leave–One–Out method, the results were very similar, between 72% and 82% classification accuracy. There was also a decrease in entropy of glycaemia records during the time interval studied. These findings provide a solid foundation to assess how glucose metabolism may be influenced by DJBL implantation and opens a new line of research in this field.

Topography of emotional valence and arousal within the motor part of the subthalamic nucleus in Parkinson's

  • DOI: 10.1038/s41598-019-56260-x
  • Odkaz: https://doi.org/10.1038/s41598-019-56260-x
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Clinical motor and non-motor effects of deep brain stimulation (DBS) of the subthalamic nucleus (STN) in Parkinson's disease (PD) seem to depend on the stimulation site within the STN. We analysed the effects of the position of the stimulation electrode within the motor STN on subjective emotional experience, expressed as emotional valence and arousal ratings to pictures representing primary rewards and aversive fearful stimuli in 20 PD patients. Patients’ ratings from both aversive and erotic stimuli matched the mean ratings from a group of 20 control subjects at similar position within the STN. Patients with electrodes located more posteriorly reported both valence and arousal ratings from both the rewarding and aversive pictures as more extreme. Moreover, posterior electrode positions were associated with a higher occurrence of depression at a long-term follow-up. This brain–behavior relationship suggests a complex emotion topography in the motor part of the STN. Both valence and arousal representations overlapped and were uniformly arranged anterior-posteriorly in a gradient-like manner, suggesting a specific spatial organization needed for the coding of the motivational salience of the stimuli. This finding is relevant for our understanding of neuropsychiatric side effects in STN DBS and potentially for optimal electrode placement.

Characterization of Artifact Influence on the Classification of Glucose Time Series Using Sample Entropy Statistics

  • Autoři: Cuesta, D., doc. Ing. Daniel Novák, Ph.D., Burda, V., Molina-Pico, A., Vargas, B., Mraz, M., Kavalkova, P., Benes, P., Haluzik, M.
  • Publikace: Entropy. 2018, 20(871), ISSN 1099-4300.
  • Rok: 2018
  • DOI: 10.3390/e20110871
  • Odkaz: https://doi.org/10.3390/e20110871
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    This paper analyses the performance of SampEn and one of its derivatives, Fuzzy Entropy (FuzzyEn), in the context of artifacted blood glucose time series classification. This is a difficult and practically unexplored framework, where the availability of more sensitive and reliable measures could be of great clinical impact. Although the advent of new blood glucose monitoring technologies may reduce the incidence of the problems stated above, incorrect device or sensor manipulation, patient adherence, sensor detachment, time constraints, adoption barriers or affordability can still result in relatively short and artifacted records, as the ones analyzed in this paper or in other similar works. This study is aimed at characterizing the changes induced by such artifacts, enabling the arrangement of countermeasures in advance when possible. Despite the presence of these disturbances, results demonstrate that SampEn and FuzzyEn are sufficiently robust to achieve a significant classification performance, using records obtained from patients with duodenal-jejunal exclusion. The classification results, in terms of area under the ROC of up to 0.9, with several tests yielding AUC values also greater than 0.8, and in terms of a leave-one-out average classification accuracy of 80%, confirm the potential of these measures in this context despite the presence of artifacts, with SampEn having slightly better performance than FuzzyEn.

eHealth Intervention for Smoking Cessation for Czech Tobacco Smokers: Pilot Study of User Acceptance

  • Autoři: Kulhánek, A., Gabrhelík, R., doc. Ing. Daniel Novák, Ph.D., Burda, V., Brendryen, H.
  • Publikace: Adiktologie. 2018, 2018(2), 81-85. ISSN 1213-3841.
  • Rok: 2018
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    BACKGROUND: The development of information and communication technologies is bringing new therapeutic options, including behavioural changes in the area of health promotion. The eHealth interventions also offer new options in efforts to stop smoking. AIM: A pilot study aimed at the assessment of the functionality aspects and user acceptance of an eHealth application for quitting smoking. METHODS: The study used a mixedmethods design and was conducted from July 2016 to February 2017. We recruited 34 tobacco users (solely cigarette smokers). Thirty respondents tested the eHealth application on their own mobile devices for a predefined period of time (up to one month). Quantitative data was collected with a data management system of the eHealth program. User acceptance was surveyed through structured telephone interviews. Feedback from the users was collected via qualitative focus groups. Quantitative analysis was performed using descriptive statistics; qualitative data was analysed with the cluster analysis method. RESULTS: The respondents completed 10 days of the pre-quitting phase on average and three weeks of the quitting phase, with a total of 19 delivered and completed online sessions. Overall, the therapeutic aspects of the eHealth intervention were seen as positive. Nearly 75% of all the actively participating respondents (n=30) preferred the eHealth intervention to seeking other professional services during the quitting phase. The study confirmed the acceptance of the new treatment modality from the point of view of the target group of tobacco smokers, despite some technical issues accompanying the pilot launch of the intervention. The eHealth application that was evaluated constitutes a promising and innovative direction in addiction treatment.

Fusion of microelectrode neuronal recordings and MRI landmarks for automatic atlas fitting in deep brain stimulation surgery

  • DOI: 10.1007/978-3-030-01201-4_19
  • Odkaz: https://doi.org/10.1007/978-3-030-01201-4_19
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    The deep brain stimulation (DBS) is a symptomatic treatment technique used mainly for movement disorders, consisting of chronic electrical stimulation of subcortical structures. To achieve very precise electrode implantation, which is necessary for a good clinical outcome, many surgical teams use electrophysiological recording around the assumed target, planned in pre-operative MRI images. In our previous work, we developed a probabilistic model to fit a 3D anatomical atlas of the subthalamic nucleus to the recorded microelectrode activity in Parkinson’s disease (PD) patients. In this paper, we extend the model to incorporate characteristic landmarks of the target nucleus, manually annotated in pre-operative MRI data. We validate the approach on a set of 27 exploration five-electrode trajectories from 15 PD patients. The results show that such combined approach may lead to a vast improvement in optimization reliability, while maintaining good fit to the electrophysiology data. The combination of electrophysiology and MRI-based data thus provides a promising approach for compensating brain shift, occuring during the surgery and achieving accurate localization of recording sites in DBS surgery.

Sample Entropy Analysis of Noisy Atrial Electrograms during Atrial Fibrillation

  • Autoři: Maria Cirugeda-Roldan, E., Molina Pico, A., doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Křemen, V.
  • Publikace: Computational and Mathematical Methods in Medicine. 2018, 2018 ISSN 1748-6718.
  • Rok: 2018
  • DOI: 10.1155/2018/1874651
  • Odkaz: https://doi.org/10.1155/2018/1874651
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Most cardiac arrhythmias can be classified as atrial flutter, focal atrial tachycardia, or atrial fibrillation. They have been usually treated using drugs, but catheter ablation has proven more effective. This is an invasive method devised to destroy the heart tissue that disturbs correct heart rhythm. In order to accurately localise the focus of this disturbance, the acquisition and processing of atrial electrograms form the usual mapping technique. They can be single potentials, double potentials, or complex fractionated atrial electrogram (CFAE) potentials, and last ones are the most effective targets for ablation. The electrophysiological substrate is then localised by a suitable signal processing method. Sample Entropy is a statistic scarcely applied to electrograms but can arguably become a powerful tool to analyse these time series, supported by its results in other similar biomedical applications. However, the lack of an analysis of its dependence on the perturbations usually found in electrogram data, such as missing samples or spikes, is even more marked. This paper applied SampEn to the segmentation between non-CFAE and CFAE records and assessed its class segmentation power loss at different levels of these perturbations. The results confirmed that SampEn was able to significantly distinguish between non-CFAE and CFAE records, even under very unfavourable conditions, such as 50% of missing data or 10% of spikes.

Feature subset selection and classification of intracardiac electrograms during atrial fibrillation

  • Autoři: Duque, S.I., Orozco-Duque, A., Křemen, V., doc. Ing. Daniel Novák, Ph.D., Tobón, C., Bustamante, J.
  • Publikace: Biomedical Signal Processing and Control. 2017, 38 182-190. ISSN 1746-8094.
  • Rok: 2017
  • DOI: 10.1016/j.bspc.2017.06.005
  • Odkaz: https://doi.org/10.1016/j.bspc.2017.06.005
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Several approaches have been adopted for the identification of arrhythmogenic sources maintaining atrial fibrillation (AF). In this paper, we propose a classifier that discriminates between four classes of atrial electrogram (EGM). We delved into the relation between levels of fractionation in EGM signals and the fibrillation substrates in simulated episodes of chronic AF. Several feature extraction methods were used to calculate 92 features from 429 real EGM records acquired during radiofrequency ablation of chronic AF. We selected the optimal subset of features by using a genetic algorithm, followed by K-nearest neighbors (K-NN) classification into four levels of fractionation. Sensitivity of 0.90 and specificity of 0.97 were achieved. Subsequently, the results of the classification were extrapolated to signals of a 3D human atria model and a 2D model of atrial tissue. The 3D model simulated an episode of AF maintained by a rotor in the posterior wall of the left atrium and the 2D model simulated an AF episode with one stable rotor. We used the K-NN classifier trained on a given set of real EGM signals to detect a specific class of signals presenting the highest level of fractionation located near the rotor's vortex. This method needs to be tested on real clinical data to provide evidence that it can support ablation therapy procedures.

Methods for automatic detection of artifacts in microelectrode recordings

  • DOI: 10.1016/j.jneumeth.2017.07.012
  • Odkaz: https://doi.org/10.1016/j.jneumeth.2017.07.012
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    BACKGROUND: Extracellular microelectrode recording (MER) is a prominent technique for studies of extracellular single-unit neuronal activity. In order to achieve robust results in more complex analysis pipelines, it is necessary to have high quality input data with a low amount of artifacts. We show that noise (mainly electromagnetic interference and motion artifacts) may affect more than 25% of the recording length in a clinical MER database. NEW METHOD: We present several methods for automatic detection of noise in MER signals, based on (i) unsupervised detection of stationary segments, (ii) large peaks in the power spectral density, and (iii) a classifier based on multiple time- and frequency-domain features. We evaluate the proposed methods on a manually annotated database of 5735 ten-second MER signals from 58 Parkinson's disease patients. COMPARISON WITH EXISTING METHODS: The existing methods for artifact detection in single-channel MER that have been rigorously tested, are based on unsupervised change-point detection. We show on an extensive real MER database that the presented techniques are better suited for the task of artifact identification and achieve much better results. RESULTS: The best-performing classifiers (bagging and decision tree) achieved artifact classification accuracy of up to 89% on an unseen test set and outperformed the unsupervised techniques by 5–10%. This was close to the level of agreement among raters using manual annotation (93.5%). CONCLUSION: We conclude that the proposed methods are suitable for automatic MER denoising and may help in the efficient elimination of undesirable signal artifacts.

Endocrine effects of duodenal-jejunal exclusion in obese patients with type 2 diabetes mellitus

  • Autoři: Kavalkova, P., Mraz, M., Trachta, P., Klouckova, J., Burda, V., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Journal of endocrinology. 2016, 231(1), 11-22. ISSN 0022-0795.
  • Rok: 2016
  • DOI: 10.1530/JOE-16-0206
  • Odkaz: https://doi.org/10.1530/JOE-16-0206
  • Pracoviště: Analýza a interpretace biomedicínských dat
  • Anotace:
    Duodenal-jejunal bypass liner (DJBL) is an endoscopically implantable device designed to noninvasively mimic the effects of gastrointestinal bypass operations by excluding the duodenum and proximal jejunum from the contact with ingested food. The aim of our study was to assess the influence of DJBL on anthropometric parameters, glucose regulation, metabolic and hormonal profile in obese patients with type 2 diabetes mellitus (T2DM) and to characterize both the magnitude and the possible mechanisms of its effect. Thirty obese patients with poorly controlled T2DM underwent the implantation of DJBL and were assessed before and 1, 6 and 10 months after the implantation, and 3 months after the removal of DJBL. The implantation decreased body weight, and improved lipid levels and glucose regulation along with reduced glycemic variability. Serum concentrations of fibroblast growth factor 19 (FGF19) and bile acids markedly increased together with a tendency to restoration of postprandial peak of GLP1. White blood cell count slightly increased and red blood cell count decreased throughout the DJBL implantation period along with decreased ferritin, iron and vitamin B12 concentrations. Blood count returned to baseline values 3 months after DJBL removal. Decreased body weight and improved glucose control persisted with only slight deterioration 3 months after DJBL removal while the effect on lipids was lost. We conclude that the implantation of DJBL induced a sustained reduction in body weight and improvement in regulation of lipid and glucose. The increase in FGF19 and bile acids levels could be at least partially responsible for these effects.

Evaluation of diabetes mellitus compensation after one year of using Mobiab system

  • DOI: 10.1109/EMBC.2016.7592096
  • Odkaz: https://doi.org/10.1109/EMBC.2016.7592096
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper shows how the Mobiab system is useful for patients and which features are usually used. 18 users of the Mobiab system have been chosen. The data for this study was collected since February 2015 to February 2016. Both patients suffering from diabetes mellitus and non-diabetic users of the Mobiab system participated in the study. The results indicate that for more than a half of users is the Mobiab system convenient and used most of all functionalities of the system. One third of the users use only limited subset of functionalities - e.g. glucose and insulin monitoring.

Metabolomic profiling of urinary changes in mice with monosodium glutamate-induced obesity

  • Autoři: Pelantová, H., Bártová, S., Ing. Jiří Anýž, Ph.D., Holubová, M., Železná, B., Maletínská, L., doc. Ing. Daniel Novák, Ph.D., Lacinová, Z., Šulc, M., Haluzík, M., Kuzma, M.
  • Publikace: Analytical and Bioanalytical Chemistry (print version). 2016, 2016(Volume 48), 567-578. ISSN 1618-2642.
  • Rok: 2016
  • DOI: 10.1007/s00216-015-9133-0
  • Odkaz: https://doi.org/10.1007/s00216-015-9133-0
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Obesity with related complications represents a widespread health problem. The etiopathogenesis of obesity is often studied using numerous rodent models. The mouse model of monosodium glutamate (MSG)-induced obesity was exploited as a model of obesity combined with insulin resistance. The aim of this work was to characterize the metabolic status of MSG mice by NMR-based metabolomics in combination with relevant biochemical and hormonal parameters. NMR analysis of urine at 2, 6, and 9 months revealed altered metabolism of nicotinamide and polyamines, attenuated excretion of major urinary proteins, increased levels of phenylacetylglycine and allantoin, and decreased concentrations of methylamine in urine of MSG-treated mice. Altered levels of creatine, citrate, succinate, and acetate were observed at 2 months of age and approached the values of control mice with aging. The development of obesity and insulin resistence in 6-month-old MSG mice was also accompanied by decreased mRNA expressions of adiponectin, lipogenetic and lipolytic enzymes and peroxisome proliferator-activated receptor- gamma in fat while mRNA expressions of lipogenetic enzymes in the liver were enhanced. At the age of 9 months, biochemical parameters of MSG mice were normalized to the values of the controls. This fact pointed to a limited predictive value of biochemical data up to age of 6 months as NMR metabolomics confirmed altered urine metabolic composition even at 9 months.

Probabilistic model of neuronal background activity in deep brain stimulation trajectories

  • DOI: 10.1007/978-3-319-43949-5_7
  • Odkaz: https://doi.org/10.1007/978-3-319-43949-5_7
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a probabilistic model for classification of micro- EEG signals, recorded during deep brain stimulation surgery for Parkinson’s disease. The model uses parametric representation of neuronal background activity, estimated using normalized root-mean-square of the signal. Contrary to existing solutions using Bayes classifiers or Hidden Markov Models, our model uses smooth state-transitions represented by sigmoid functions, which ensures flexible model structure in combination with general optimizers for parameter estimation and model fitting. The presented model can easily be extended with additional parameters and constraints and is intended for fitting of a 3D anatomical model to micro-EEG data in further perspective. In an evaluation on 260 trajectories from 61 patients, the model showed classification accuracy 90.0%, which was comparable to existing solutions. The evaluation proved the model successful in target identification and we conclude that its use for more complex tasks in the area of DBS planning and modeling is feasible.

Characterization of Complex Fractionated Atrial Electrograms by Sample Entropy: An International Multi-Center Study

  • Autoři: Cirugeda–Roldán, E., doc. Ing. Daniel Novák, Ph.D., Křemen, V., Cuesta–Frau, D., Keller, M., Luik, A., Šrutová, M.
  • Publikace: Entropy. 2015, 17(11), 7493-7509. ISSN 1099-4300.
  • Rok: 2015
  • DOI: 10.3390/e17117493
  • Odkaz: https://doi.org/10.3390/e17117493
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Atrial fibrillation (AF) is the most commonly clinically-encountered arrhythmia. Catheter ablation of AF is mainly based on trigger elimination and modification of the AF substrate. Substrate mapping ablation of complex fractionated atrial electrograms (CFAEs) has emerged to be a promising technique. To improve substrate mapping based on CFAE analysis, automatic detection algorithms need to be developed in order to simplify and accelerate the ablation procedures. According to the latest studies, the level of fractionation has been shown to be promisingly well estimated from CFAE measured during radio frequency (RF) ablation of AF. The nature of CFAE is generally nonlinear and nonstationary, so the use of complexity measures is considered to be the appropriate technique for the analysis of AF records. This work proposes the use of sample entropy (SampEn), not only as a way to discern between non-fractionated and fractionated atrial electrograms (A-EGM), but also as a tool for characterizing the degree of A-EGM regularity, which is linked to changes in the AF substrate and to heart tissue damage. The use of SampEn combined with a blind parameter estimation optimization process enables the classification between CFAE and non-CFAE with statistical significance (p < 0:001), 0.89 area under the ROC, 86% specificity and 77% sensitivity over a mixed database of A-EGM combined from two independent CFAE signal databases, recorded during RF ablation of AF in two EU countries (542 signals in total). On the basis of the results obtained in this study, it can be suggested that the use of SampEn is suitable for real-time support during navigation of RF ablation of AF, as only 1.5 seconds of signal segments need to be analyzed.

Distinct populations of neurons respond to emotional valence and arousal in the human subthalamic nucleus

  • Autoři: Mgr. Tomáš Sieger, Ph.D., Serranová, T., Růžička, F., Vostatek, P., Wild, J., Šťastná, D., Bonnet, C., doc. Ing. Daniel Novák, Ph.D., Růžička, E., Urgošík, D., Jech, R.
  • Publikace: Proceedings of the National academy of sciences of the United Stated of America. 2015, 112(10), 3116-3121. ISSN 0027-8424.
  • Rok: 2015
  • DOI: 10.1073/pnas.1410709112
  • Odkaz: https://doi.org/10.1073/pnas.1410709112
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Both animal studies and studies using deep brain stimulation in humans have demonstrated the involvement of the subthalamic nucleus (STN) in motivational and emotional processes; however, participation of this nucleus in processing human emotion has not been investigated directly at the single-neuron level. We analyzed the relationship between the neuronal firing from intraoperative microrecordings from the STN during affective picture presentation in patients with Parkinsons disease (PD) and the affective ratings of emotional valence and arousal performed subsequently. We observed that 17% of neurons responded to emotional valence and arousal of visual stimuli according to individual ratings. The activity of some neurons was related to emotional valence, whereas different neurons responded to arousal. In addition, 14% of neurons responded to visual stimuli. Our results suggest the existence of neurons involved in processing or transmission of visual and emotional information in the human STN, and provide evidence of separate processing of the affective dimensions of valence and arousal at the level of single neurons as well.

Mobiab System for Diabetes Mellitus Compensation

  • Autoři: Burda, V., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: INTERNATIONAL WORKSHOP ON COMPUTATIONAL INTELLIGENCE FOR MULTIMEDIA UNDERSTANDING. Praha: Czech Technical University in Prague, 2015. ISBN 978-1-4673-8457-5.
  • Rok: 2015
  • DOI: 10.1109/IWCIM.2015.7347078
  • Odkaz: https://doi.org/10.1109/IWCIM.2015.7347078
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The Mobiab system is complex system for diabetes mellituscompensation. This system is designed as client-server withmobile application and web portal. The application cancollect data about patient’s caloric intake and expenditure,can communicate via Bluetooth with glucose meter,tonometer and scales. The mobile application is usinggamification principles for patient's positive motivation. Theweb portal gives to physicians a way to supervise theirpatients. The Mobiab system is almost prepared for pilottesting and after successful test also for clinical study.

Multifractal analysis for grading complex fractionated electrograms in atrial fibrillation

  • Autoři: Orozco-Duque, A., doc. Ing. Daniel Novák, Ph.D., Křemen, V., Bustamante, J.
  • Publikace: Physiological Measurement. 2015, 34(11), 2269-2284. ISSN 0967-3334.
  • Rok: 2015
  • DOI: 10.1088/0967-3334/36/11/2269
  • Odkaz: https://doi.org/10.1088/0967-3334/36/11/2269
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Complex fractionated atrial electrograms provide an important tool for identifying arrhythmogenic substrates that can be used to guide catheter ablation for atrial fibrillation (AF). However, fractionation is a phenomenon that remains unclear. This paper aims to evaluate the multifractal properties of electrograms in AF in order to propose a method based on multifractal analysis able to discriminate between different levels of fractionation. We introduce a new method, the h-fluctuation index (hFI), where h is the generalised Hurst exponent, to extract information from the shape of the multifractal spectrum. Two multifractal frameworks are evaluated: multifractal detrended fluctuation analysis and wavelet transform modulus maxima. hFI is exemplified through its application in synthetic signals, and it is evaluated in a database of electrograms labeled on the basis of four degrees of fractionation. We compare the performance of hFI with other indexes, and find that hFI outperforms them. The results of the study provide evidence that multifractal analysis is useful for studying fractionation phenomena in AF electrograms, and indicate that hFI can be proposed as a tool for grade fractionation associated with the detection of target sites for ablation in AF.

Optimization of Parkinson Disease Treatment Combining Anti-Parkinson Drugs and Deep Brain Stimulation Using Patient Diaries

  • DOI: 10.1109/EMBC.2015.7319133
  • Odkaz: https://doi.org/10.1109/EMBC.2015.7319133
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The number of patients suffering from Parkinson’s disease is increasing rapidly due to population aging. While traditional medications-based palliative therapy is successful in early stages, deep brain stimulation (DBS) may be used as an alternative treatment in later stages. After DBS implantation, the therapy typically consists of electrical stimulation and reduced medication. In order to provide good clinical outcome, a balance has to be found between medication and stimulation parameters, this is usually done as follows: First, Unified Parkinson's Disease Rating Scale (UPDRS) scoring is performed, second patients are supposed to fill subjective diaries during a specific period. This study shows that these diaries are useful as therapy progression indicator. Feel scores based on diaries and sleep time were examined with respect to DBS stimulation and medication. The results confirmed the positive effect of both therapy components - stimulation as well as medication - on patient feel scores. Furthermore, a positive correlation was observed between stimulation energy and sleep duration.

Perioperative tight glucose control reduces postoperative adverse events in non-diabetic cardiac surgery patients

  • Autoři: Bláha, J., Mráz, M., Kopecký, P., Stříteský, M., Lipš, M., Matias, M., Kunstýř, J., Pořízka, M., Kotulák, T., Kolníková, I., Šimanovská, B., Zakharchenko, M., Rulíšek, J., Šachl, R., Ing. Jiří Anýž, Ph.D., doc. Ing. Daniel Novák, Ph.D., Lindner, J., Hovorka, R., Svačina, Š., Haluzík, M.
  • Publikace: The Journal of Clinical Endocrinology & Metabolism. 2015, Volume 100(Issue 8), 3081-3089. ISSN 1945-7197.
  • Rok: 2015
  • DOI: 10.1210/jc.2015-1959
  • Odkaz: https://doi.org/10.1210/jc.2015-1959
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Context: Tight glucose control (TGC) reduces morbidity and mortality in patients undergoing elective cardiac surgery, but only limited data about its optimal timing are available to date. Objective: To compare the effects of perioperative (PERI) versus postoperative (POST) initiation of TGC on postoperative adverse events in cardiac surgery patients. Design: Single center, single-blind, parallel-group, randomized controlled trial. Settings: Academic tertiary hospital. Participants: 2383 hemodynamically stable patients undergoing major cardiac surgery with expected postoperative ICU treatment for at least 2 consecutive days. Intervention: Perioperatively or postoperatively initiated intensive insulin therapy with target glucose range 4.4–6.1 mmol/l. Main Outcome Measures: Adverse events from any cause during postoperative hospital stay. Results: In the whole cohort, perioperatively initiated TGC markedly reduced the number of postoperative complications (23.2 vs. 34.1%, 95% CI 0.60–0.78) in spite of only minimal improvement in glucose control (blood glucose 6.6±0.7 vs. 6.7±0.8 mmol/l, p<0.001; time in target range 39.3±13.7 vs. 37.3±13.8%, p<0.001). The positive effects of TGC on postoperative complications were driven by non-diabetic subjects (21.3 vs. 33.7%, 95% CI 0.54–0.74; blood glucose 6.5±0.6 vs. 6.6±0.8 mmol/, n.s.; time in target range 40.8±13.6 vs. 39.7±13.8%, n.s.), while no significant effect was seen in diabetic patients (29.4 vs. 35.1%, 95% CI 0.66–1.06) despite significantly better glucose control in the PERI group (blood glucose 6.9±1.0 vs. 7.1±0.8 mmol/l, p<0.001; time in target range 34.3±12.7 vs. 30.8±11.5%, p<0.001). Conclusions: Perioperative initiation of intensive insulin therapy during cardiac surgery reduces postoperative morbidity in non-diabetic patients while having minimal effect in diabetic subjects.

Strategy for NMR metabolomic analysis of urine in mouse models of obesity–from sample collection to interpretation of acquired data

  • DOI: 10.1016/j.jpba.2015.06.036
  • Odkaz: https://doi.org/10.1016/j.jpba.2015.06.036
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The mouse model of monosodium glutamate induced obesity was used to examine and consequently optimize the strategy for analysis of urine samples by NMR spectroscopy. A set of nineteen easily detectable metabolites typical in obesity-related studies was selected. The impact of urine collection protocol, choice of 1H NMR pulse sequence, and finally the impact of the normalization method on the detected concentration of selected metabolites were investigated. We demonstrated the crucial effect of food intake and diurnal rhythms resulting in the choice of a 24-hour fasting collection protocol as the most convenient for tracking obesity-induced increased sensitivity to fasting. It was shown that the Carr-Purcell-Meiboom-Gill (CPMG) experiment is a better alternative to 1D-NOESY for NMR analysis of mouse urine due to its ability to filter undesirable signals of proteins naturally present in rodent urine. Normalization to total spectral area provided comparable outcomes as did normalization to creatinine or probabilistic quotient normalization in the CPMG-based model. The optimized approach was found to be beneficial mainly for low abundant metabolites rarely monitored due to their overlap by strong protein signals.

Strong Identification and Authentication Using Dynamic Biometric Signature

  • DOI: 10.1007/978-3-662-45402-2_175
  • Odkaz: https://doi.org/10.1007/978-3-662-45402-2_175
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper discusses the possibility of usage of dynamic biometric signatures (DBS) as an alternative of “advanced” electronic signatures. The authors refer to the issues of uniqueness of DBS in terms of individual’s characteristic motion, and hence the validity of parameters required for methods of identification of users and their authentication in the “electronic” world. Furthermore the authors focused on the issue of the use of DBS in connection with legal acts of people and the integration of authentication tools, based on DBS in the legal systems of the EU countries. The conclusions set out in the paper are supported by the results of experiments, when panelists created their biometric samples under different conditions that may have a significant impact, especially for set-up setting of FNMR parameter (False Non-Match Rate).

Supervised Segmentation of Microelectrode Recording Artifacts Using Power Spectral Density

  • DOI: 10.1109/EMBC.2015.7318661
  • Odkaz: https://doi.org/10.1109/EMBC.2015.7318661
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Appropriate detection of clean signal segments in extracellular microelectrode recordings (MER) is vital for maintaining high signal-to-noise ratio in MER studies. Existing alternatives to manual signal inspection are based on unsupervised change-point detection. We present a method of supervised MER artifact classification, based on power spectral density (PSD) and evaluate its performance on a database of 95 labelled MER signals. The proposed method yielded test-set accuracy of 90%, which was close to the accuracy of annotation (94%). The unsupervised methods achieved accuracy of about 77% on both training and testing data.

The dynamic biometric signature - is the biometric data in the created signature constant?

  • DOI: 10.1109/CCST.2015.7389715
  • Odkaz: https://doi.org/10.1109/CCST.2015.7389715
  • Pracoviště: Katedra fyziky, Analýza a interpretace biomedicínských dat
  • Anotace:
    Biometric authentication methods are increasingly proving to be a sensible compromise between the demands on the user and/or the authentication tools without reducing the level of security. Dynamic biometric signature (DBS) systems record data from the handwritten signature using special tools which enable an analysis of both the static and dynamic properties associated with the typical behaviour of the signing individual.

The Dynamic Biometric Signature. Is the Biometric Data in the Created Signature Constant?

  • Pracoviště: Katedra fyziky, Katedra kybernetiky
  • Anotace:
    Biometric authentication methods are increasingly proving to be a sensible compromise between the demands on the user and/or the authentication tools without reducing the level of security. Dynamic biometric signature (DBS) systems record data from the handwritten signature using special tools which enable an analysis of both the static and dynamic properties associated with the typical behaviour of the signing individual.

Alternative User Interface for Blind Users

  • Autoři: Svobodník, P., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2014, ISBN 978-1-4244-7929-0.
  • Rok: 2014
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper deals with design and implementation of user interface which makes accessible environment of the mobile operation system Android 4.0 and above for visually impaired users (especially for blind). Interface enables to perform basic operations with the system and common used touch gestures. Voice synthesizer, vibration and sound are implemented as a feedback. Interface was tested with 12 target users following man-machine design process including interviews with focus group and tests in a usability laboratory.

Analysis of Actigraph Parameters for Relapse Prediction in Bipolar Disorder: A Feasibility Study

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Albert, F., Španiel, F.
  • Publikace: Proceedings of 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2014. p. 4972-4975. ISBN 978-1-4244-7929-0.
  • Rok: 2014
  • DOI: 10.1109/EMBC.2014.6944740
  • Odkaz: https://doi.org/10.1109/EMBC.2014.6944740
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The paper presents a framework for early iden- tification of prodromal syndromes od mania or depression in bipolar disorder. The framework may mitigate relapses and improve patient functioning. The methodology consists of long- term actigraphy monitoring and simplified self-assessment tool to determine manic or depression events. Eight patients were involved in the feasibility study, spanning period of 150 months, resulting in 17 relapses and 3 hospitalizations in total. We concluded that the most promising parameter extracted from actigraphy recording is a circadian rhythm’s interdaily stability Using developed trend analysis applied on interdaily stability parameter, we achieved sensitivity and specificity about 65, resp. 68. We hypothesized that this performance is both mainly due to missing values in data and due to small amount of relapses.

Atrial Electrogram Complex Fractionated Entropy Study

  • Autoři: Roldán, E.M.C., Picó, A.M., doc. Ing. Daniel Novák, Ph.D., Frau, D.C., Křemen, V.
  • Publikace: Experimental & Clinical Cardiology. 2014, 20(9), 5566-5574. ISSN 1205-6626.
  • Rok: 2014
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Atrial electrogram (AEGM) signal processing started to play an important role during radiofrequency ablation (RFA) of atrial fibrillation (AF). This work uses Sample Entropy (SampEn) as a tool for quantification and evaluation of AEGM complexity and discrimination between complex fractionated AEGM (CFAE) and non CFAE. Additionally, SampEn classification capabilities and its robustness are tested against the presence of spikes and sample loss. Both signal disturbances were superimposed to the original AEGM signals. AEGM signals were measured during RFA of AF in patients recommended for RFA of AF. The database consisted of 113 AEGM signals of short length, 1.5 s, classified by experts into nonCFAE and CFAE classes. The results demonstrated that SampEn conserved its clinical validity up to a 10% of spike contamination (Cxy > 0.8) even though it was able to differentiate between CFAE and nonCFAE classes up to a spike probability of 0.5. It also proved to differentiate between nonCFAE and CFAE classes with a data loss percentage of up to 50%. In conclusion, SampEn is a low computational cost, easy to implement, and robust tool to characterize and quantify complexity of AEGM in real time.

Diabetes Mellitus Compensation Using Mobile Technology

  • Autoři: Burda, V., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of the 2nd Conference on Mobile and Information Technologies in Medicine. Praha: České vysoké učení technické v Praze, Fakulta elektrotechnická, 2014, ISBN 978-80-01-05637-0. Available from: http://mobmed.org/download/proceedings2014/mobileMed2014_paper_27.pdf
  • Rok: 2014
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We are working on system for diabetes mellitus compensation - Mobiab project. The whole system is designed as client-server with mobile application and web portal. The application can collect data about patient’s caloric intake and expenditure, can communicate via Bluetooth with glucose meter, tonometer and scales. The application is using gamification principles for patient's positive motivation. The web portal gives physicians way to supervise their patients. The Mobiab project is almost prepared for pilot testing and after successful test also for clinical study.

Dynamic Approximate Entropy Electroanatomic Maps Detect Rotors in a Simulated Atrial Fibrillation Model

  • Autoři: Ugarte, J.P., Orozco-Duque, A., Tobón, C., Křemen, V., doc. Ing. Daniel Novák, Ph.D., Saiz, J., Oesterlein, C., Schmitt, C., Luik, A., Bustamante, J.
  • Publikace: PLoS ONE. 2014, (9)12(9(12)), 1-19. ISSN 1932-6203.
  • Rok: 2014
  • DOI: 10.1371/journal.pone.0114577
  • Odkaz: https://doi.org/10.1371/journal.pone.0114577
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    There is evidence that rotors could be drivers that maintain atrial fibrillation. Complex fractionated atrial electrograms have been located in rotor tip areas. However, the concept of electrogram fractionation, defined using time intervals, is still controversial as a tool for locating target sites for ablation. We hypothesize that the fractionation phenomenon is better described using non-linear dynamic measures, such as approximate entropy, and that this tool could be used for locating the rotor tip. The aim of this work has been to determine the relationship between approximate entropy and fractionated electrograms, and to develop a new tool for rotor mapping based on fractionation levels. Two episodes of chronic atrial fibrillation were simulated in a 3D human atrial model, in which rotors were observed. Dynamic approximate entropy maps were calculated using unipolar electrogram signals generated over the whole surface of the 3D atrial model. In addition, we optimized the approximate entropy calculation using two real multi-center databases of fractionated electrogram signals, labeled in 4 levels of fractionation. We found that the values of approximate entropy and the levels of fractionation are positively correlated. This allows the dynamic approximate entropy maps to localize the tips from stable and meandering rotors. Furthermore, we assessed the optimized approximate entropy using bipolar electrograms generated over a vicinity enclosing a rotor, achieving rotor detection. Our results suggest that high approximate entropy values are able to detect a high level of fractionation and to locate rotor tips in simulated atrial fibrillation episodes. We suggest that dynamic approximate entropy maps could become a tool for atrial fibrillation rotor mapping.

Feature Selection for Discrimination of Fractionation Levels in Atrial Electrograms

  • Autoři: Orozco-Duque, A., Martínez-Vargas, J.D., doc. Ing. Daniel Novák, Ph.D., Bustamante, J., Castellanos-Dominguez, G.
  • Publikace: Proceedings of 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2014, pp. 1595-1598. ISBN 978-1-4244-7929-0.
  • Rok: 2014
  • DOI: 10.1109/EMBC.2014.6943909
  • Odkaz: https://doi.org/10.1109/EMBC.2014.6943909
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Radiofrequency catheter ablation of atrial fibrillation (AF) guided by complex fractionated atrial electrograms (CFAE) is associated with a high AF termination rate in paroxysmal AF, but not in persistent. CFAE does not always identify favorable sites for persistent AF ablation. Studies suggest that only high fractionation level should be used as a target site for ablation. Nonetheless, there are not a standardized criterion to defined fractionation levels. Therefore, a better characterization of the signal is required providing a set of more powerful features that should be extracted from CFAE. Due to the apparent difference among fractionation classes in terms of their stochastic variability, we test time-domain and time-frequency based feature extraction approaches. Also, we carried out the symmetrical uncertainty-based feature selection to determine the most relevant features which improve discrimination of fractionation levels. Obtained results on a tested real electrogram database show that most relevant features in time-domain are related with time intervals and not with amplitudes. Nonetheless, time-frequency features obtained more information from the signal and this representation is likely a better suitable discriminating approach, particularly to detect high fractionated electrograms with a sensitivity and specificity of 83.0% and 93.6%, respectively.

Fractionated Electrograms and Rotors Detection in Chronic Atrial Fibrillation Using Model-Based Clustering

  • Autoři: Orozco-Duque, A., Duque, S.I., Ugarte, J.P., Tobon, C., doc. Ing. Daniel Novák, Ph.D., Křemen, V., Castellanos-Dominguez, G., Saiz, J., Bustamante, J.
  • Publikace: Proceedings of 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2014. p. 1579-1582. ISBN 978-1-4244-7929-0.
  • Rok: 2014
  • DOI: 10.1109/EMBC.2014.6943905
  • Odkaz: https://doi.org/10.1109/EMBC.2014.6943905
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The identification of atrial fibrillation (AF) substrates is needed to improve ablation therapy guided by electrograms (EGM), although mechanisms that sustain AF are not fully understood. Detection of complex fractionated atrial electrograms (CFAE) is used for this purpose. Nonetheless, efficacy of this method is poor in the case of chronic AF. Recent hypothesis proposes the rotors as fibrillatory substrate. Novel approaches seek to relate CFAE with rotor; nevertheless, such methods are not able to identify the associated substrate. Furthermore, the patterns that characterize CFAE generated by rotors remain unknown. Thus, tracking of rotors is an unsolved issue. In this paper we propose a non-supervised method to find patterns associated with fibrillatory substrates in chronic AF. We extracted two features based on local activation wave detection and one feature based on non-linear dynamics. Gaussian mixture model-based clustering was used to discriminate CFAE patterns. Resulting clusters were visualized in an electroanatomic map. We assessed the proposed method in a real database labeled according to the level of fractionation and in a simulated episode of chronic AF in which a rotor was detected. Our results indicate that the method proposed is able to separate different levels of fractionation in CFAE, and provide evidence that clustering can be used to locate the vortex of the rotors. Such method can aid ablation therapy procedures by means of CFAE patterns discrimination.

GLYCEMIC VARIABILITY DOES NOT PREDICT POSTOPERATIVE OUTCOMES IN ELECTIVE CARDIAC SURGERY PATIENTS ON TIGHT GLUCOSE CONTROL

  • Autoři: Mraz, M., Kopecky, P., Lips, M., doc. Ing. Daniel Novák, Ph.D., Lindner, J., Svacina, S., Blaha, J., Haluzik, M.
  • Publikace: Diabetes Technology and Therapeutics. 2014, 16(1), A6. ISSN 1520-9156.
  • Rok: 2014
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    GLYCEMIC VARIABILITY DOES NOT PREDICT POSTOPERATIVE OUTCOMES IN ELECTIVE CARDIAC SURGERY PATIENTS ON TIGHT GLUCOSE CONTROL RELATIONSHIP BETWEEN GLYCEMIC VARIABILITY, GLUCOSE CONTROL AND LONG-TERM COMPLICATIONS IN PATIENTS WITH DIABETES MELLITUS

The Influence of Deep Hypothermia on Inflammatory Status, Tissue Hypoxia and Endocrine Function of Adipose Tissue During Cardiac Surgery

  • Autoři: Drapalova, J., Kopecky, P., Bartlova, M., Lacinova, Z., doc. Ing. Daniel Novák, Ph.D., Maruna, P., Lips, M., Mraz, M., Lindner, J., Haluzik, M.
  • Publikace: Cryobiology. 2014, 68(2), 269-275. ISSN 0011-2240.
  • Rok: 2014
  • DOI: 10.1016/j.cryobiol.2014.02.007
  • Odkaz: https://doi.org/10.1016/j.cryobiol.2014.02.007
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Changes in endocrine function of adipose tissue during surgery, such as excessive production of proinflammatory cytokines, can significantly alter metabolic response to surgery and worsen its outcomes and prognosis of patients. Therapeutic hypothermia has been used to prevent damage connected with perioperative ischemia and hypoperfusion. The aim of our study was to explore the influence of deep hypothermia on systemic and local inflammation, adipose tissue hypoxia and adipocytokine production. We compared serum concentrations of proinflammatory markers (CRP, IL-6, IL-8, sIL-2R, sTNFRI, PCT) and mRNA expression of selected genes involved in inflammatory reactions (IL-6, TNF-α, MCP-1, MIF) and adaptation to hypoxia and oxidative stress (HIF1-α, MT3, GLUT1, IRS1, GPX1, BCL-2) in subcutaneous and visceral adipose tissue and in isolated adipocytes of patients undergoing cardiosurgical operation with hypothermic period. Deep hypothermia significantly delayed the onset of surgery-related systemic inflammatory response. The relative gene expression of the studied genes was not altered during the hypothermic period, but was significantly changed in six out of ten studied genes (IL-6, MCP-1, TNF-α, HIF1-α, GLUT1, GPX1) at the end of surgery. Our results show that deep hypothermia suppresses the development of systemic inflammatory response, delays the onset of local adipose tissue inflammation and thus may protect against excessive expression of proinflammatory and hypoxia-related factors in patients undergoing elective cardiac surgery procedure.

Basal Ganglia Neuronal Activity during Scanning Eye Movements in Parkinson’s Disease

  • DOI: 10.1371/journal.pone.0078581
  • Odkaz: https://doi.org/10.1371/journal.pone.0078581
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The oculomotor role of the basal ganglia has been supported by extensive evidence, although their role in scanning eye movements is poorly understood. Nineteen Parkinsońs disease patients, which underwent implantation of deep brain stimulation electrodes, were investigated with simultaneous intraoperative microelectrode recordings and single channel electrooculography in a scanning eye movement task by viewing a series of colored pictures selected from the International Affective Picture System. Four patients additionally underwent a visually guided saccade task. Microelectrode recordings were analyzed selectively from the subthalamic nucleus, substantia nigra pars reticulata and from the globus pallidus by the WaveClus program which allowed for detection and sorting of individual neurons. The relationship between neuronal firing rate and eye movements was studied by crosscorrelation analysis. Out of 183 neurons that were detected, 130 were found in the subthalamic nucleus, 30 in the substantia nigra and 23 in the globus pallidus. Twenty percent of the neurons in each of these structures showed eye movement-related activity. Neurons related to scanning eye movements were mostly unrelated to the visually guided saccades. We conclude that a relatively large number of basal ganglia neurons are involved in eye motion control. Surprisingly, neurons related to scanning eye movements differed from neurons activated during saccades suggesting functional specialization and segregation of both systems for eye movement control.

Case Studies of Students´ Involvement in Research

  • Autoři: Lhotská, L., Kužílek, J., Chudáček, V., Novák, P., doc. Ing. Daniel Novák, Ph.D., Ing. Jan Havlík, Ph.D.,
  • Publikace: Proceedings of the 24th Annual Conference on Proceedings of the 24th Annual Conference on European Association for Education in Proceedings of the 24th Annual Conference on European Association for Education in p Electrical and Information Engineering. Piscataway: IEEE, 2013, pp. 204-209. ISBN 978-1-4799-0043-5. Available from: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6576530&punumber%3D6573516%26sortType%3Dasc_p_Sequence%26filter%3DAND%28p_IS_Number%3A6576484%29%26pageNumber%3D2
  • Rok: 2013
  • DOI: 10.1109/EAEEIE.2013.6576530
  • Odkaz: https://doi.org/10.1109/EAEEIE.2013.6576530
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Research and development have been speeding up in recent decades, especially in interdisciplinary and newly emerging areas. Since the students will be exposed to this situation immediately after graduation when they start their jobs it is desirable to involve them in research projects. Through this work they acquire new knowledge and skills they will need in the future jobs. Moreover, they are frequently participating in strongly interdisciplinary research that requires orientation not only in engineering disciplines, but also in the other concerned areas. One of the typical examples is the area of Biomedical Engineering. In this paper we will discuss several case studies of students´ involvement in research and present their research results.

Diaphragm Postural Function Analysis Using Magnetic Resonance Imaging

  • Autoři: Vostatek, P., doc. Ing. Daniel Novák, Ph.D., Rychnovský, T., Rychnovská, Š.
  • Publikace: PLoS ONE. 2013, 8(3), 1-13. ISSN 1932-6203.
  • Rok: 2013
  • DOI: 10.1371/journal.pone.0056724
  • Odkaz: https://doi.org/10.1371/journal.pone.0056724
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a postural analysis of diaphragm function using magnetic resonance imaging (MRI). The main aim of the study was to identify changes in diaphragm motion and shape when postural demands on the body were increased (loading applied to a distal part of the extended lower extremities against the flexion of the hips was used). Sixteen healthy subjects were compared with 17 subjects suffering from chronic low back pain and in whom structural spine disorders had been identified. Two sets of features were calculated from MRI recordings: dynamic parameters reflecting diaphragm action, and static parameters reflecting diaphragm anatomic characteristics. A statistical analysis showed that the diaphragm respiratory and postural changes were significantly slower, bigger in size and better balanced in the control group. When a load was applied to the lower limbs, the pathological subjects were mostly not able to maintain the respiratory diaphragm function, which was lowered significantly. Subjects from the control group showed more stable parameters of both respiratory and postural function. Our findings consistently affirmed worse muscle cooperation in the low back pain population subgroup. A clear relation with spinal findings and with low back pain remains undecided, but various findings in the literature were confirmed. The most important finding is the need to further address various mechanisms used by patients to compensate deep muscle insufficiency.

Does IT Bring Hope for Wellbeing?

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Štěpánková, O., Rousseaux, S., Busuoli, M., Carulli, M., D´Agosta, G., Gallelli, T., Uller, M., Mráz, M., Haluzík, M., Richter, P., Martin, M., Wilson, R., Bettocchi, M., Mucci, T., Pipitone, E., Lamberti, C., Siena, A., Descovich, C., Dvořáček, B., Petioky, Ch., Babič, F., Lenart, M., Wöhrer, A.
  • Publikace: Handbook of Research on ICTs for Human-Centered Healthcare and Social Care Services (2 Volumes). Hershey, Pennsylvania: IGI Global, 2013. p. 270-302. ISBN 978-1-4666-3986-7.
  • Rok: 2013
  • DOI: 10.4018/978-1-4666-3986-7.ch014
  • Odkaz: https://doi.org/10.4018/978-1-4666-3986-7.ch014
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The first part of this chapter reviews the design, implementation, and customer experience with the OLDES SW tele-care platform developed within the EU project Older people’s e-services at home. The OLDES solution has been successfully tested at two different locations: in Italy with the participation of a group of 100 seniors (including 10 senior citizens suffering from heart disease), and in the Czech Republic, with the involvement of a group of 10 diabetic patients. The suggested OLDES approach proved to be an effective solution for municipalities, hospitals, and their contact centres for providing health and social services. The project partners therefore decided to develop a second generation of the system called SPES (Support to Patients through E-Service Solutions), which started in April 2011. The SPES project aims at transferring the original approach and results achieved in implementing the OLDES focusing on new target problem domains: dementia, mobility-challenged persons, respiratory problems, and social exclusion.

Hidden Markov Models for Analysis of Eye Movements of Dyslexic Children

  • Autoři: Macaš, M., Lhotská, L., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of the 18th International Conference on Digital Signal Processing. Piscataway: IEEE, 2013, ISSN 1546-1874. ISBN 978-1-4673-5805-7. Available from: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6622783
  • Rok: 2013
  • DOI: 10.1109/ICDSP.2013.6622783
  • Odkaz: https://doi.org/10.1109/ICDSP.2013.6622783
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The paper describes an application of hidden Markov models to dyslexia detection from eye movements. Eye movements of reading-age dyslexic and control children are measured, pre-processed and hidden Markov model with two hidden states is trained on velocity time series for each child. The two states of the model correspond to two component of the eye movements signal - fixations and saccades. The elements of transition matrix are further used one by one as features for 1-dimensional linear Bayes classifier. It is shown that this method applied to eye movements during the simplest non-verbal task can lead to relatively high performance. Thus, we propose this feature extraction for a more sophisticated systems which would be able to detect dyslexia in pre-school children.

Methods for Students´Motivation During the Biomedical Engineering Study

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Promoting the students motivation is surely an integral part of the university education. It enhances their interest in the study branch and increases the probability of finishing the class not to say the whole university education. In our contribution, we discuss different methods which were used in innovated classes at our university. The methods include working on a common project, using real data, increasing course interdisciplinarity. making software assignments more attractive by implementation into a robotic device and applicability of solved tasks in the commercial sphere. We focus in more detail on results of one of our projects, which aimed at development of a toolbox for sample psychological experiments, synchronized with an EEG device. Thanks to the fact that the whole project is using modular architecture, students can implement individual modules during their semestral, bachelor or diploma theses or can further extend functionality of the toolbox itself.

Serum Preadipocyte Factor-1 Concentrations in Females with Obesity and Type 2 Diabetes Mellitus: The Influence of Very Low Calorie Diet, Acute Hyperinsulinemia, and Fenofibrate Treatment

  • Autoři: Kaválková, P., Toušková, V., Roubíček, T., Trachta, P., Urbanová, M., Drápalová, J., Haluzíková, D., Mráz, M., doc. Ing. Daniel Novák, Ph.D., Matoulek, M., Lacinová, Z., Haluzík, M.
  • Publikace: HORMONE AND METABOLIC RESEARCH. 2013, 45(11), 820-826. ISSN 0018-5043.
  • Rok: 2013
  • DOI: 10.1055/s-0033-1353210
  • Odkaz: https://doi.org/10.1055/s-0033-1353210
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Appropriate differentiation capacity of adipose tissue significantly affects its ability to store lipids and to protect nonadipose tissues against lipid spillover and development of insulin resistance. Preadipocyte factor-1 (Pref-1) is an important negative regulator of preadipocyte differentiation. The aim of our study was to explore the changes in circulating Pref-1 concentrations in female subjects with obesity (OB) (n=19), females with obesity and type 2 diabetes mellitus (T2DM) (n=22), and sex- and age-matched healthy control subjects (C) (n=22), and to study its modulation by very low calorie diet (VLCD), acute hyperinsulinemia during isoglycemic-hyperinsulinemic clamp, and 3 months' treatment with PPAR- agonist fenofibrate. At baseline, serum Pref-1 concentrations were significantly higher in patients with T2DM compared to control group, while only nonsignificant trend towards higher levels was observed in OB group. 3 weeks of VLCD decreased Pref-1 levels in both OB and T2DM group, whereas 3 months of fenofibrate treatment had no significant effect. Hyperinsulinemia during the clamp significantly suppressed Pref-1 levels in both C and T2DM subjects and this suppression was unaffected by fenofibrate treatment. In a combined population of all groups, circulating Pref-1 levels correlated positively with insulin, leptin and glucose levels and HOMA (homeostasis model assessment) index. We conclude that elevated Pref-1 concentrations in T2DM subjects may contribute to impaired adipose tissue differentiation capacity associated with insulin resistance in obese patients with T2DM. The decrease of Pref-1 levels after VLCD may be involved in the improvement of metabolic status and the amelioration of insulin resistance in T2DM patients.

SPES Services Deliver Hope for People Suffering from Dementia

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Uller, M., Štěpánková, O., Petioky, Ch., Kolouchová, A.
  • Publikace: Proceedings of the AAL Forum 2013. Eindhoven: Smart Homes, 2013,
  • Rok: 2013
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    SPES (Support patients through E-services solutions) project implements telehealth and social platform focusing on four pilot project: dementia, respiratory problems, handicapped people and social exclusion. We present a technological ecosystem for people suffering from dementia in early stages. The main objectives are reminiscence therapy andsupporting localization of clients outside home. The first module of the SPES platform is a web based multimedia applications that allows to create and view a personal multimedia memorial booklet with any number of sides and individually selected content placed invarios avaible layouts. The application can be used in two modes. In the online mode, the application is hosted on the server where the media files are recorded and wherepossible using a web interface memorial book not only view, but also create and edit (creator mode for books). The resulting book can also be exported as a zip archivecantaining all the files that make up the book and even the browser, after unpacking the archivem the book can be viewed without access to the server (offline mode). The useof application is supported by a cargiver whose role is to fill multimedia content in according to client needs. The second module of the SPES platform in an mobile Android applicationenabling localization of a client with accompanying web interface. While people moving in an open area are monitored using GPS technologym WiFi networks are used in the confined spaces. A nurse/carer can set the borders through the web interface. As soon as the client leaves the area, the care-giver receives a warning SMS. Naturally, the client´s location can be visualized through the website as well. The SPES platform for dementia is evaluated in day care centres in Vienna and Brno. Preliminary results indicate very fast acceptance of both modules by carresponding users.

Estimation of Respiratory Parameters from the Photoplethysmogram

  • Autoři: Ing. Eduard Bakštein, Ph.D., Strašrybka, T., Štěpánková, O., Macků, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: BioDat 2012 - Conference on Advanced Methods of Biological Data and Signal Processing. Praha: České vysoké učení technické v Praze, 2012, ISBN 978-80-01-05153-5. Available from: http://bio.felk.cvut.cz/biodat/BioDat%202012_conference_program.pdf
  • Rok: 2012
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Photoplethysmography (PPG) is a non-invasive low-cost investigation method, suitable for long-term monitoring of various physiological properties, such as heart rate or blood oxygen saturation. This paper describes method of estimation of respiratory parameters from the PPG signal. In our preliminary study, PPG signal was recorded along with air intake on 8 healthy subjects. Two algorithms, using signal filtering and envelope detection for respiratory parameter estimation are described and evaluated. Relative breath depth was estimated from the recorded PPG signal and compared to actual parameters of the subject's breath, measured by a spirometer device.

Performance Comparison of Extracellular Spike Sorting Algorithms for Single-Channel Recordings

  • DOI: 10.1016/j.jneumeth.2011.10.013
  • Odkaz: https://doi.org/10.1016/j.jneumeth.2011.10.013
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Proper classification of action potentials from extracellular recordings is essential for making an accurate study of neuronal behavior. Many spike sorting algorithms have been presented in the technical literature. However, no comparative analysis has hitherto been performed. In our study, three widely-used publicly-available spike sorting algorithms (WaveClus, KlustaKwik, OSort) were compared with regard to their parameter settings. The algorithms were evaluated using 112 artificial signals (publicly available online) with 2-9 different neurons and varying noise levels between 0.00 and 0.60.

Pro koho je SPES nadějí?

  • Autoři: Štěpánková, O., doc. Ing. Daniel Novák, Ph.D., Novák, P., Uller, M., Mráz, M., Haluzík, M.
  • Publikace: Sborník příspěvků MEDSOFT 2012. Praha: Creative Connections, 2012, pp. 267-282. ISSN 1803-8115. Available from: http://spes-project.eu/doc/MedSoft2012SPES.pdf
  • Rok: 2012
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Projekt SPES (Supporting Patients through E-service Solutions, 2011 - 2013) staví na praktických zkušenostech získaných v rámci projektu OLDES a snaží se zobecnit jeho metodiku tak, aby výsledný systém byl aplikovatelný pro různé typy klientů domácí péče. Vývoj a testování navržené koncepce e-handicap je tématem čtyř pilotních studií, z nichž jedna je věnována podpoře sociálně vyloučených seniorů (Košice) a zbylé tři skupiny mají zřejmý telemedicínský charakter: Italská Ferrara se zaměří na pacienty s chronickými dýchacími chorobami, především na ty, kteří potřebují neinvazivní podporu ventilace a kyslíkovou terapii. Ve Vídni, se budou věnovat pacientům s demencí, pro které jsou navrhovány a testovány nejen programy na podporu a trénink paměti, ale i systém lokalizace ztracených osob v budovách a současně v otevřeném prostoru. Důraz je kladen na uživatelsky příjemné prostředí, které uvažovaní klienti budou schopni využívat. V Boskovicích na Moravě má systém sloužit imobilním pacientům, kterým nejen zpřístupní služby v oblasti e-health, ale současně nabídne možnost ověřovat v běžném denním provozu, jak jsou pro ně skutečně užitečné různé aktuálně dostupné technické pomůcky, např. senzory identifikace nebezpečných sklonů, místní dálkové ovládání domácnosti (zapnutí / vypnutí domácích spotřebičů, oken a dveří, atd.) nebo identifikační náramky pro dezorientované osoby. Příspěvek nejprve představí projekty OLDES i SPES, shrne dosud získané zkušenosti a zmíní se o aktuálním stavu řešení projektu SPES. V závěru se pokusí upozornit na problémy, které bude třeba řešit, aby systémy podobného charakteru mohly být dlouhodobě v činnosti a skutečně se tak do budoucna stát velkou nadějí pro zlepšení kvality života seniorů, nemocných i těch, kteří o ně pečují.

Wrapper Feature Selection for Small Sample Size Data Driven by Complete Error Estimates

  • DOI: 10.1016/j.cmpb.2012.02.006
  • Odkaz: https://doi.org/10.1016/j.cmpb.2012.02.006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper focuses on wrapper-based feature selection for a 1-nearest neighbor classifier. We consider in particular the case of a small sample size with a few hundred instances, which is common in biomedical applications. We propose a technique for calculating the complete bootstrap for a 1-nearest-neighbor classifier (i.e., averaging over all desired test/train partitions of the data). The complete bootstrap and the complete cross-validation error estimate with lower variance are applied as novel selection criteria and are compared with the standard bootstrap and cross-validation in combination with three optimization techniques - sequential forward selection (SFS), binary particle swarm optimization (BPSO) and simplified social impact theory based optimization (SSITO). The experimental comparison based on ten datasets draws the following conclusions: for all three search methods examined here, the complete criteria are a significantly better choice than standard 2-fold cross-validation, 10-fold cross-validation and bootstrap with 50 trials irrespective of the selected output number of iterations. All the complete criterion-based 1NN wrappers with SFS search performed better than the widely-used FILTER and SIMBA methods. We also demonstrate the benefits and properties of our approaches on an important and novel real-world application of automatic detection of the subthalamic nucleus.

Neuronal Activity of the Basal Ganglia and Subthalamus in Relation to Eye Movement in Parkinson's Disease

  • Autoři: Mgr. Tomáš Sieger, Ph.D., Bonnet, C., Serranová, T., Wild, J., doc. Ing. Daniel Novák, Ph.D., Růžička, F., Urgošík, D., Jech, R.
  • Publikace: Abstracts of the 14th European Congress of Clinical Neurophysiology and the 4th International Conference on Transcranial Magnetic and Direct Current Stimulation, Clinical Neurophysiology. Dublin: Elsevier Irland Ltd., 2011, pp. s89-s90. ISSN 1388-2457. Available from: http://www.sciencedirect.com/science/journal/13882457/122/supp/S1
  • Rok: 2011
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The basal ganglia and the thalamus are involved in eye motion (EM) control. It has been shown that deep brain stimulation (DBS) of the subthalamic nucleus (STN) affects EM in patients with Parkinson's disease (PD), and that the STN contains neurons activated during voluntary saccades or pursuit EM. Our results suggested that each of the explored structures - STN, SNr and GPi contain relatively high ratios of neurons involved in the execution and/or control of eye movements.

Automatic Nuclei Detection During Parkinson's Stereotactic Neurosurgery

  • Autoři: Wild, J., doc. Ing. Daniel Novák, Ph.D., Ing. Eduard Bakštein, Ph.D., Jech, R.
  • Publikace: Analysis of Biomedical Signals and Images, BIOSIGNAL 2010, Proceedings. Brno: Brno University of Technology, 2010, pp. 48-49. ISSN 1211-412X. ISBN 978-80-214-4106-4. Available from: http://www.biosignal.cz/bs2010/papers/1108.pdf
  • Rok: 2010
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Although single-cell microelectrode recordings (MER) are used to confirm stereotactic targets during surgery for movement disorders, there is a lack of automatic exploration methods which are designed for guiding surgeon during identifying appropriate target for deep-brain stimulation (DBS) implant. We propose automatic visualization method for MER to determine corresponding deep brain nuclei. The underground hypothesis is that nuclei automatic identification can help to determine the subthalamic nucleus (STN) in Parkinson's disease patients. This approach aims at improving patient outcome by helping neurosurgeons objectively identify target structures.

Detekční algoritmy pro klasifikaci neuronální aktivity

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Jeden z hlavních problémů současné neuroinformatiky je plně automaticky ale přesto spolehlivě extrahovat a roztřídit akční potenciály z extracelularního záznamu. Tímto problémem se však zabývá hned několik algoritmů, přičemž každý je potřeba před samotným použitím nakonfigurovat s ohledem na druh zpracovávaného záznamu. Provedli jsme nezávislé výkonnostní porovnání tří nejčastěji používaných algoritmů - KlustaKwik, WaveClus a OSort - a doplnili jej o metodiku optimalizace jejich nastavení.

Diaphragm Postural Function Analysis Using Magnetic Resonance

  • Autoři: Vostatek, P., doc. Ing. Daniel Novák, Ph.D., Rychnovský, T., Wild, J.
  • Publikace: 10th International Conference on Information Technology and Applications in Biomedicine. Crete: IEEE Control Syst Soc, 2010, pp. 99. ISBN 978-1-4244-6560-6.
  • Rok: 2010
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Comparison of changes of diaphragm motion and shape for 16 healthy subjects are compared to 17 subjects suffering from chronic pain and having structural spine findings

Neuronální aktivita bazálních ganglií vázaná na oční pohyby u pacientů s parkinsonovou nemocí

  • Autoři: Mgr. Tomáš Sieger, Ph.D., Wild, J., doc. Ing. Daniel Novák, Ph.D., Jech, R., Bonnet, C., Seranová, T., Růžička, F., Urgošík, D.
  • Publikace: 57. SPOLEČNÝ SJEZD ČESKÉ A SLOVENSKÉ SPOLEČNOSTI PRO KLINICKOU NEUROFYZIOLOGII. Praha: MH Consulting, 2010, pp. 11.
  • Rok: 2010
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Hluboké mozková stimulace (DBS) subthalamického jádra (STN) v léčbě Parkinsonovy nemoci může způsobovat paréza očních pohybů a diplopii. Vyhledávali jsem proto neurony, jejichž aktivita by souvisela s vykonáváním očních pohybů. Neuronální aktivitu jsme kromě v STN analyzovali také v substantia nigra (SNr) a ve vnitřním pallidu (GPi). Výsledky naznačují, že jádra STN, SNr a GPi obsahují vysoký počet neuronů, jejichž aktivita souvisí s vykonáváním očních pohybů.

Diabetes Management in OLDES Project

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Uller, M., Rousseaux, S., Mráz, M., Smrž, J., Štěpánková, O., Haluzík, M., Busuoli, M.
  • Publikace: Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2009. p. 7228-7231. ISSN 1557-170X. ISBN 978-1-4244-3295-0.
  • Rok: 2009
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    EU project OLDES (Older People's e-services at home) develops easy to use and low cost ICT platform in order to offer a better quality of life to elderly people directly in their homes through innovative systems of teleaccompany, tele-assistance and tele-medicine. The elderly are able to access the services and send relevant medical data from their home by being connected to the central server via a low cost PC which is based on Negroponte paradigm. The OLDES platform interface uses television screens controlled through a remote control customized for the elderly. The feasibility of OLDES project is evaluated by the pilot study concentrating on compensation of diabetic patients. Compensation of diabetes is achieved by monitoring glucose glycemia level, blood pressure and weight. Moreover, the patient feeds into OLDES system daily consumption of food using interactive food scales and obtains advice if necessary.

Discrimination of Endocardial Electrogram Disorganization Using a Signal Regularity Analysis

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Křemen, V., Cuesta-Frau, D., Chudáček, V., Lhotská, L.
  • Publikace: Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2009. p. 1812-1815. ISSN 1557-170X. ISBN 978-1-4244-3295-0.
  • Rok: 2009
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Measures from the theory of nonlinear dynamics were applied on complex fractionated atrial electrograms (CFAEs) in order to characterize their physiological dynamic behavior. The results were obtained considering 113 short term atrial electrograms (A-EGMs) which were annotated by three experts into four classes of fractionation according to A-EGMs signal regularity. The following measures were applied on AEGM signals: General Correlation Dimension, Approximate Entropy, Detrended Fluctuation Analysis, Lempel-Ziv Complexity, and Katz-Sevcik, Variance and Box Counting Fractal Dimension. Assessment of disorganization was evaluated by a Kruskal Wallis statistical test. Except Detrended Fluctuation Analysis and Variance Fractal Dimension, the CFAE disorganization was found statistically significant even for low significant level α = 0.001. Moreover, the increasing complexity of A-EGM signals was reflected by higher values of General Correlation Dimension of order 1 and Approximate Entro

Identifying Number of Neurons in Extracellular Recording

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    One of the most difficult aspects of spike sorting is choosing the number of neurons in extracellular recording. The paper proposes a methodology for estimating the number of neurons based on the Gaussian mixture model. The following criteria have been examined: Bayesian selection method, Akaikes information criteria, minimum description length, minimum message length, fuzzy hyper volume, evidence density and partition coefficient. In order to validate the procedure, an experimental comparative study was carried out, comparing the proposed methodology with three spike sorting algorithms. The proposed methodology has an advantage of setting the minimum number of parameters and is very robust to background noise. We conclude that only fuzzy hyper volume and evidence density criteria are able to identify the correct number of neurons across different noise levels.

Measuring Body Temperature Series Regularity Using Approximate Entropy and Sample Entropy

  • Autoři: Cuesta-Frau, D., Miró-Martínez, M., Oltra-Crespo, S., Varela-Entrecanales, M., Aboy, M., doc. Ing. Daniel Novák, Ph.D., Austin, D.
  • Publikace: Proceedings of the 31st Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway: IEEE, 2009, pp. 3461-3464. ISSN 1557-170X. ISBN 978-1-4244-3295-0.
  • Rok: 2009
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Approximate Entropy (ApEn) and Sample Entropy (SampEn) have proven to be a valuable analyzing tool for a number of physiological signals. However, the characterization of these metrics is still lacking. We applied ApEn and SampEn to body temperature time series recorded from patients in critical state. This study was aimed at finding the optimal analytical configuration to best distiguishing between survivor and non-survivor records, and at gaining additional insight into the characterization of such tools. A statistical analysis of the results was conducted to support the parameter and metric selection criteria for this type of physiological signals.

Analysis of Human Brain NMR Spectra in Vivo Using Artificial Neural Networks

  • Autoři: Saudek, E., doc. Ing. Daniel Novák, Ph.D., Wágnerová, D., Hájek, M.
  • Publikace: Artificial Neural Networks - ICANN 2008. Heidelberg: Springer, 2008, pp. 517-526. ISSN 0302-9743. ISBN 978-3-540-87558-1.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Magnetic resonance has proven to be a successful method of in-vivo imaging. Although MRI can help detect various pathologies, its ability to classify the nature of the pathological tissue is limited. Magnetic resonance spectroscopy allows identifying metabolite content of the tissue and estimating the metabolite concentration. Map of metabolite concentration along with the MR image allows proper classification of many pathologies, for example progressive tumorous tissue identification in brain. Standard methods used to analyze nuclear magnetic resonance spectra such as singular value decomposition or curve fitting algorithms are very time consuming taking several minutes to analyze spectrum from a single voxel. To analyze the spectra from a chemical shift imagine sequence (CSI) in maximal resolution hundreds of spectra need to be processed. The suggested ANN framework proved to be much faster. Networks were trained on the outputs of LCModel curve fitting algorithm.

Analysis of Vestibular-Ocular Reflex by Evolutionary Framework

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Pilný, A., Kordík, P., Holiga, Š., Ing. Petr Pošík, Ph.D., Černý, R., Brzezný, R.
  • Publikace: Artificial Neural Networks - ICANN 2008, PT I. Heidelberg: Springer, 2008. pp. 452-461. ISSN 0302-9743. ISBN 978-3-540-87535-2.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper the problem of analysis of eye movements using sinusoidal head rotation test is presented. The goal of the method is to discard automatically the effect of the fast phase-saccades and consequently calculate the response of vestibular system in the form of phase shift and amplitude. The comparison of threshold detection and inductive models trained on saccades is carried out. After saccades detection we are left with discontinuous signal segments. This paper presents an approach to align them to form a smooth signal with the same frequencies that were originally present in the source signal. The approach is based on a direct estimation of the signal component parameters using the evolutionary strategy with covariance matrix adaptation. The performance of evolutionary approach is compared to least-square multimodal sinus fit.

Better diabetes compensation in OLDES project

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Štěpánková, O., Mráz, M., Haluzík, M.
  • Publikace: Analysis of Biomedical Signals and Images; Biosignal 2008 proceedings. Brno: VUTIUM Press, 2008, ISSN 1211-412X. ISBN 978-80-214-3613-8.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    EU project OLDES aims at developing a very low cost and easy to use entertainment and health care platform designed to ease the life of older people in their homes. The platform is based on a PC corresponding to Negroponte's paradigm of a € 100 device. OLDES combines user entertainment services (through easy-to-access thematic interactive channels and special interest forums supported by animators) and health care facilities.

ITAREPS Improves Life Quality of Schizophrenic Patients

  • Autoři: Hrdlička, J., doc. Ing. Daniel Novák, Ph.D., Španiel, F.
  • Publikace: Efektivita, kvalita a spokojenost klientů ve zdravotnictví a sociální péči. Pardubice: STAPRO s.r.o., 2008, pp. 42. ISBN 978-80-903879-1-1.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The patients with psychotic diseases, especially with have one of the highest relapse rates characterized by long-stay hospitalization. The system ITAREPS (Information Technology Aided Relapse Prevention in Schizophrenia) has been developed for rapid and targeted recognition of early warning signs of psychotic disorder relapse. The goal of this work is to design and evaluate an algorithm for recognition of schizophrenia and bipolar symptoms and for prediction of sudden illness exacerbation and patient re-hospitalization.

ITAREPS: Information Technology Aided Relapse Prevention Programme in Schizophrenia

  • Autoři: Španiel, F., Vohlídka, P., Hrdlička, J., Kožený, J., Novák, T., Motlová, L., Čermák, J., Bednařík, J., doc. Ing. Daniel Novák, Ph.D., Höschl, C.
  • Publikace: Schizophrenia Research. 2008, 98(1-3), 312-317. ISSN 0920-9964.
  • Rok: 2008
  • DOI: 10.1016/j.schres.2007.09.005
  • Odkaz: https://doi.org/10.1016/j.schres.2007.09.005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    ITAREPS presents a mobile phone-based telemedicine solution for weekly remote patient monitoring and disease management in schizophrenia and psychotic disorders in general. The programme provides health professionals with home telemonitoring via a PC-to-phone SMS platform that identifies prodromal symptoms of relapse, to enable early intervention and prevent unnecessary hospitalizations. Its web-based interface offers the authorized physician a longitudinal analysis of the dynamics and development of possible prodromes.

Long Short-Term Memory for Apnea Detection based on Heart Rate Variability

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Mucha, K., Al-ani, T.
  • Publikace: Proceedings of the 30th Annual International Conference on the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society, 2008. p. 5234-5237. ISSN 1557-170X. ISBN 978-1-4244-1815-2.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The main drive force in apnea current diagnostic is to reduce overwhelming number of sleep disorders candidates by means of very simple-to-use, comfortable and cheap methodology. The proposed framework is based only on automatic analysis of electrocardiogram signal. The feature extraction stage was performed using methods of Heart Rate Variability and Detrended Fluctuation analysis. Feature-spaces formed using these two methods were used as input to a Long Short-Term Memory Artificial Neural Network chosen for its capability to find temporally dependencies in the data. The framework was evaluated on Challenge 2000 Physionet database yielding successful rate 82.1%, sensitivity 85.5% and specificity 80.1%.

OLDES: Improved Welfare for Diabetic Patients

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Štěpánková, O., Mráz, M., Haluzík, M.
  • Publikace: Efektivita, kvalita a spokojenost klientů ve zdravotnictví a sociální péči. Pardubice: STAPRO s.r.o., 2008, pp. 43. ISBN 978-80-903879-1-1.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    High standards of welfare in developed world has led to overall aging of population thus increasing the percentage of people in need of supportive care at home. The OLDES project of tele-accompany approach represents a step forward of the EU funded research project giving aged people living at home a digital companion, in the form of a low cost PC. The device will implement various entertainment channels, but in a way that makes information transparent and easy to use for the old and disabled people. The approach proposed is an integrated web-based system able to manage information and "knowledge" regarding the domestic environment and the biological conditions.

OLDES: new solution for long-term diabetes compensation management

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Štěpánková, O., Mráz, M., Haluzík, M., Bussuoli, M., Uller, M., Malý, K., Nováková, L., Novák, P.
  • Publikace: Proceedings of the 30th Annual International Conference on the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society, 2008. pp. 4346-4349. ISSN 1557-170X. ISBN 978-1-4244-1815-2.
  • Rok: 2008
  • DOI: 10.1109/IEMBS.2008.4650172
  • Odkaz: https://doi.org/10.1109/IEMBS.2008.4650172
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Abstract-EU project OLDES (Older People's e-services at home) aims at developing a very low cost and easy to use entertainment and health care platform designed to ease the life of older people in their homes. The platform is based on a PC corresponding to Negroponte's paradigm of a 100 $ device. OLDES combines user entertainment services (through easy-to-access thematic interactive channels and special interest forums supported by animators) and health care facilities. The pilot case study of diabetes type II compensation under the OLDES framework is presented. Apart from measurement of continuous glucose, blood pressure and weight, the user feeds into OLDES system food daily consumption using interactive food scales via user friendly software interface designed by user-centered design paradigm and obtains advice if necessary.

Particle Swarm Optimization of Hidden Markov Models: a comparative study

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Macaš, M.
  • Publikace: Distributed Human-Machine Systems. Praha: CTU Publishing House, 2008, pp. 235-239. ISBN 978-80-01-04027-0.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In recent years, Hidden Markov Models (HMM) have been increasingly applied in data mining applications. However, most authors have used classical optimization Expectation- Maximization (EM) scheme. A new method of HMM learning based on Particle Swarm Optimization (PSO) has been developed. Along with others global approaches as Simulating Annealing (SIM) and Genetic Algorithms (GA) the following local gradient methods have been also compared: classical Expectation-Maximization algorithm, Maximum A Posteriory approach (MAP) and Bayes Variational learning (VAR). The methods are evaluated on a synthetic data set using different evaluation criteria including classification problem. The most reliable optimization approach in terms of performance, numerical stability and speed is VAR learning followed by PSO approach.

Postural Function Analysis using Propriomed

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Vostatek, P., Rašav, E.
  • Publikace: Analysis of Biomedical Signals and Images; Biosignal 2008 proceedings. Brno: VUTIUM Press, 2008, ISSN 1211-412X. ISBN 978-80-214-3613-8.
  • Rok: 2008
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Wrong body posture is the source of 95% pain in human locomotive system. Postural function keeps body position against gravitation. Postural functions can be improved by a series of exercise using therapeutic tool Propriomed. The main goal of the proposed study is to objectively evaluate data obtained from rehabilitation process based on series of Propriomed exercises.

Automatic Sleep Scoring Based Only on Electrocardiogram Records

  • Autoři: Al-ani, T., Kazbunda, R., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of the 6th EUROSIM Congress on Modelling and Simulation. Vienna: ARGESIM, 2007, pp. 1-11. ISBN 978-3-901608-32-2.
  • Rok: 2007
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The technique of Polysomnography used in the diagnostic procedure for sleep studies require observation of numerous signals during sleep time. Evaluating these signals in 30 seconds interval is time consuming even for experience physician. Because that these signals are recorded in digital form and the diagnosis is made directly from these records, so they are suitable for automatic processing. The aim of this work is to automatically classify sleep stage using only the electrocardiogram (ECG) records. Our approach has been tested on a real ECG records from different patients demonstrating the feasibility of the proposed method. The capability to differentiate sleep stages in predefined categories (wake,light sleep, deep sleep, REM) was successful in 65%. The Classification performed at data set containing only deep sleep and REM categories had 83.4% reliability

Entertainment and Ambient: a New OLDES' view

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This work introduces a new concept of supporting elderly at their homes. The whole framework is being developed under OLDES project: Older People's e-services at home. OLDES aims at developing a very low cost and easy to use entertainment and health care platform designed to ease the life of older people in their homes. The platform is based on a PC corresponding to Negroponte's paradigm of a € 100 device, giving the guarantee of an affordable system. OLDES provides: user entertainment services, through easy-to-access thematic channels and special interest forums supported by animators; and health care facilities based on established Internet and tele-care communication standards. As an example of OLDES platform implementation, two pilot projects are addressed: 100 clients pilot including 10 people with cardiological problems in Bologna, Italy and diabetes pilot in Prague, Czech Republic

OLDES projekt: nová koncepce e-gerontologie

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Dlouhověkost je dnes ukazatelem civilizovanosti země. Zvyšující se životní úroveň, současná medicína a péče o vlastní zdraví jsou kreditem naší doby. S vyšším věkem se však objevují specifické zdravotní a sociální problémy. Starý člověk bývá často závislý na pomoci svého okolí, péče o něj mnohdy mění vztahy a úlohy jednotlivých členů rodiny či komunity [1]. Tato práce představuje nový koncept podpory starších lidí v prostředí domova. Celý program je zastřešen evropským projektem OLDES: Older People's e-services at home. Cílem projektu je vývoj nízkonákladového systému, který by usnadnil a obohatil život nejenom osamělým seniorům. Systém se skládá ze dvou hlavních částí: monitorovací a informační (zábavné).

A Wearable Sensor System for Recording Activity and ECG: First Results of a Clinical Application

  • Autoři: Naujokat, E., doc. Ing. Daniel Novák, Ph.D., Arndt, M., Norra, Ch.
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006, pp. 18-20. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    A wearable sensor system for recording activity and ECG: First results of a clinical application Heart Rate Variability (HRV) analysis is a widely used method for studying cardiac autonomic modulation. The purpose of the work was to determine whether by continuous monitoring of a subject's activity and HRV it is possible to determine the effect of depression treatment as depression is not only associated with altered HRV but also with elevated rates of cardiovascular morbidity and mortality. For data acquisition, a new wearable sensor system has been used.

Analysis of Vestibulo-ocular Reflex by Evolutionary Algorithm

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this paper the problem of analysis of eye movements using sinusoidal head rotation test is presented. The reflex generated by the rotational sinusoidal test is known as vestibulo-ocular reflex (VOR) producing nystagmus, which consists of slow and fast component. The goal of the method is to discard automatically the effect of the fast phase and consequently calculate the response of vestibular system in the form of phase shift and amplitude. This paper presents an approach to align the slow phases to form a smooth signal with the same frequencies that were originally present in the source signal. Two methods of direct search are compared: the Nelder-Mead simplex search and the evolutionary strategy with covariance matrix adaptation. The experimental evaluation on artificial and real-world signals revealed that the evolutionary strategy is more robust, scalable and reliable method, however, its success strongly depends on the saccades removal algorithm.

Automatic QT Interval Measurement Using Rule-Based Gradient Method

  • Autoři: Chudáček, V., Huptych, M., doc. Ing. Daniel Novák, Ph.D., Lhotská, L.
  • Publikace: Computers in Cardiology 2006. Piscataway: IEEE, 2006. ISSN 0276-6547.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In this work, we address the problem of fully automated QT-interval measurement. The method is based on detecting the Q and T points via gradient method accordingly to the beat morphology. Development and validation of the algorithm have been done using the PTB Diagnostic ECG Database prepared for Computers in Cardiology Challenge 2006. RR analysis is used for selection of the beat on which QT interval is measured. The beginning of the Q-wave is detected on all 12-leads. Considering Toff detection, morphology of repolarization phase of the beat is determined. Based on the morphology, rules are applied to particular gradients of repolarization phase to determine the T-wave end. Additional rules were applied to omit signals with high probability of the wrong QT interval measurement. With our method we have reached a score of 35.58ms.

Constraints in Particle Swarm Optimization of Hidden Markov Models

  • Autoři: Macaš, M., doc. Ing. Daniel Novák, Ph.D., Lhotská, L.
  • Publikace: Intelligent Data Engineering and Automated Learning - Proceedings of IDEAL 2006. Berlin: Springer, 2006. p. 1399-1406. ISSN 0302-9743. ISBN 3-540-45485-3.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper presents new application of Particle Swarm Optimization (PSO) algorithm for training Hidden Markov Models (HMMs). The problem of finding an optimal set of model parameters is numerical optimization problem constrained by stochastic character of HMM parameters. Constraint handling is carried out using three different ways and the results are compared to Baum-Welch algorithm (BW), commonly used for HMM training. The global searching PSO method is much less sensitive to local extremes and finds better solutions than the local BW algorithm, which often converges to local optima. The advantage of PSO approach was markedly evident, when longer training sequence was used.

Heartbeat Classification Using Gaussian Mixture Models

  • Autoři: Mico, P., Cuesta, F., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006, pp. 3-5. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Automatic analysis of long term electrocardiographic registers is an issue of great importance since these records are used to diagnose difficult to observe heart diseases and their duration makes human inspection time consuming and error prone. That we propose in this paper is a supervised procedure for heartbeat classification based on a probabilistic method successfully applied to many pattern recognition tasks, Gaussian Mixture Models (GMM). So, we will use a mixture composed by bidimensional Gaussian probability density functions (pdf) to modelate a concrete heartbeat morphology. The parameters of these Gaussian pdfs are estimated using the Expectation-Maximization (EM) algorithm, and at that point, thresholds can be obtained to separate the objects into the component classes. Experiments are carried out using registers of the MIT ECG database.

Low-cost motivated rehabilitation system for post-operation exercises

  • Autoři: Brutovský, J., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: 28th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society, 2006. ISSN 1557-170X. ISBN 1-4244-0033-3.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The incidence of joint fractures is increasing and has become one of the major health problems in developed countries. Our low-cost motivated rehabilitation system enables clinicians to prescribe, demonstrate and monitor patient rehabilitation protocols during and between clinical visits. With its unique biofeedback feature it is useful for continuous patient's motivation. The proposed system can be used in wide spectrum of rehabilitation scenario by simply downloading appropriate protocols. The hardware and software architecture (communication protocols, power management policies and application-level control) have been tuned to optimize cost, battery autonomy and real-time performance required for this application. The main advantages of the proposed system is home-based rehabilitation, low-cost and good user accetability.

New Thresholding Function for Noise Reduction in Electrocardiographic Signals

  • Autoři: Boix, M., Mora, M., Cuesta, D., Mico, P., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006, pp. 172-174. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The wavelet transform (WT) is a popular mathematical tool for signal processing nowadays. It has been used both for signal and image processing. This work is aimed at reducing the noise in electrocardiographic signals using WT and a new thresholding function. ECGs by themselves constitute an important field of application of such technique.

Particle Swarm Optimization for Hidden Markov Models with Application to Intracranial Pressure Analysis

  • Autoři: Macaš, M., doc. Ing. Daniel Novák, Ph.D., Lhotská, L.
  • Publikace: Analysis of Biomedical Signals and Images - Proceedings of Biosignal 2006. Brno: VUTIUM Press, 2006, pp. 175-177. ISSN 1211-412X. ISBN 80-214-3152-0.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The paper presents new application of Particle Swarm Optimization for training Hidden Markov Models. The approach is verified on artificial data and further, the application to Intracranial Pressure (ICP) analysis is described. In comparison with Expectation Maximization algorithm, commonly used for the HMM training problem, the PSO approach is less sensitive on sticking to local optima because of its global character. However this advantage depends on character of the particular problem. The IC analysis is the case of such problem where it is suitable to use the PSO strategy. This is demonstrated by better classification result (85.1%) in comparison with the EM algorithm (76.3%).

Unsupervised Cluster Analysis Using Particle Swarms for Oculographic Signal Segmentation

  • Autoři: Macaš, M., Lhotská, L., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: International Special Topics Conference on Information Technology in Biomedicine. Piscataway: IEEE, 2006.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The problem of real-world signal segmentation is very often difficult because of artifacts and noise. Furthermore, for each signal, a special method using supervised adaptation for every concrete type of segment must be mostly used. This paper proposes fully unsupervised approach using partitional clustering method with squared error criterion. The optimal partition is searched through the use of particle swarm optimization (PSO), which makes it possible to overcome local minima and find the near-optima solution with relatively good computational efficiency. First, the PSO clustering is tested using an artificial benchmark data set and then, practical results of the method on electrooculographic (EOG) signal segmentation are described.

Využití shlukové analýzy s částicovými hejny pro segmentaci elektrookulografického signálu

  • Autoři: Macaš, M., Lhotská, L., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: 53. společný sjezd české a slovenské společnosti klinické neurofyziologie. Praha: MH Consulting, 2006, pp. 35-36. ISBN 80-7254-916-2.
  • Rok: 2006
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Problem segmentace biologickych signalu je casto obtizny kvuli artefaktum a sumu. Navic, pro kazdy signal musi byt vetsinou pouzita zvlastni metoda adaptace pro kazdy konkretni typ segmentu. Clanek navrhuje pristup uceni bez ucitele pouzivajici shlukovani. Optimalni prirazeni do shluku je hledano pomoci optimalizace casticovymi hejny (Particle Swarm Optimization), ktera umoznuje prekonat lokalni optima a najit temer optimalni reseni v rozumnem case. Metoda je testovana na signalech ocnich pohybu namerenych elektro-okulografickou metodou.

A new data acquisition system for monitoring circadian variations of activity and ECG: Clinical application

  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Variations of circadian activity profiles and sleep patterns have proved to be altered in various psychiatric disorders as did heart rate and related parameters. For example, during a depressive episode many patients show changes in sleep, motor activity, and ECG which may be important for diagnosis as well as for treatment. Today, the evaluation of these parameters is not part of the standard diagnostic and therapeutic monitoring procedure. But they can yield valuable information, e.g. for the medication adjustment during the acute phase by assessing sleep disturbances and motor activity quantitatively and objectively.

AI+Nanotechnologies= Brave New World?

  • Autoři: Štěpánková, O., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: NANO´O5. Brno: VUT v Brně, Fakulta strojní, 2005, pp. 122. ISBN 80-214-3044-3.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Each century has its own keywords: it was steam for the 19th century, computers and information technologies for the 20th century and without any doubts it will be nanotechnology and nanomedicine for the 21st century. Nano scalpels will be able to repair cell structure, remove cancer, programmable pills equipped by nanorobots will transport and place drugs in right time on right place without any further secondary effects. It is obvious that construction of such sophisticated systems is impossible without new technical means. Each application, each specialized nanorobot will need its own unique solution - its program, i.e. a plan, which will control its activity so that it leads to the desired goal. Who will provide these programs? And how will these programs be checked for correctness? Since there will be enormous demand for this type of service, nanomedicine will not be able to achieve its goal without relying on extensive automation.

Automatic Sleep Apnea Diagnosis System Using Noninvasive Vital Signals Records

  • Autoři: Pozzo Mendoza, P., Al-ani, T., doc. Ing. Daniel Novák, Ph.D., Lhotská, L., Rigaux, L.
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005, ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Sleep Apnea Syndrome (SAS) is a very common sleep disorder. SAS is considered as clinically relevant when the breath stops during more than 10 seconds due to different factors and occurs more than five times per sleep hour. In this paper, we present an automatic approach to sleep apnea classification. This system uses only noninvasive records of the respiratory and cardiac activities (Nasal Airway Flow (NAF) and Pulse Transit Time (PTT)) issued by the technique of PolySomnoGraphy (PSG) are considered for the detection of the different sleep apnea syndromes: obstructive, central and hypopnea. Experimental results using clinical data are presented.

Bio-inspired Methods for Analysis and Classification of Reading Eye Movements of Dyslexic Children

  • Autoři: Macaš, M., Lhotská, L., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: European Symposium on Nature-inspired Smart Information Systems. 2005,
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The paper describes real practical problem of analyzing diagnostic significance of dyslexic eye movements. The biologically inspired methods were used and compared with classical methods of artificial intelligence. Eye movements of 52 school children were measured using videooculographic (VOG) technique, during a reading task. There were three groups of subjects - normal readers, retarded readers and dyslexics. The main goal was to analyze the possibility of dyslexia detection only from the eye movement signal. Time and frequency domain features were extracted and subset of significant features was chosen by a simple feature selection method. The selected feature subset was visualized using a self-organizing map (SOM). Clusters were formed by the SOM proving that proposed methodology is suitable for automatic dyslexia detection.

Clustering Improvement for Electrocardiographic Signals

  • Autoři: Micó, P., Frau, D.C., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Image Analysis and Processing - ICIAP 2005. Berlin: Springer-Verlag, 2005, pp. 892-899. ISSN 0302-9743. ISBN 3-540-28869-4.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Holter signals are ambulatory long-term electrocardiographic (ECG) registers used to detect heart diseases which are di.cult to .nd in normal ECG. These signals normally include several channels and its duration is up to 48 hours. The principal problem for the cardiologists consists of the manual inspection of the whole Holter ECG to .nd all those beats whose morphology di.er from the normal cardiac rhythm. The later analysis of these abnormal beats yields a diagnostic from the pacient's heart condition. In this paper we compare the performance among several clustering methods applied over the beats processed by Principal Component Analysis (PCA). Moreover, an outlier removing stage is added, and a cluster estimation method is included. Quality measurements, based on ECG labels from MIT-BIH database, are developed too. At the end, some results-accuracy values among several clustering algorithms is presented.

Dyslexia Detection from Eye Movements Using Artificial Neural Networks

  • Autoři: Macaš, M., Lhotská, L., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: The 3rd European Medical and Biological Engineering Conference - EMBEC´05. Praha: Společnost biomedicínského inženýrství a lékařské informatiky ČLS JEP, 2005, ISSN 1727-1983.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The main goal of the study was to propose and implement a neural network based classifier for dyslexia detection from eye movement signal. Eye movements of 76 school children were measured using videooculographic (VOG) technique during one reading and four non-reading tasks. Time and frequency domain features were extracted and various feature selection methods were performed to select subsets of significant features. Finally a feed-forward neural network using back-propagation algorithm was used for a supervised learning. A suitable topology was chose and learning parameters were set experimentally. The final classifier reached about 90% correct identification of the presence of dyslexia and about 90.5% correct identification of the absence of dyslexia.

Noninvasive Automatic Sleep Apnea Classification System

  • Autoři: Al-Ani, T., Pozzo Mendoza, P., doc. Ing. Daniel Novák, Ph.D., Hamam, Y., Lhotská, L., Lofaso, F., Isabey, D., Fodil, R.
  • Publikace: Proceedings of the Conference on Modeling and Simulation in Biology, Medicine and Biomedical Engineering. BioMedSim 2005. Linköping University, 2005, pp. 45-55.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Sleep Apnea Syndrome (SAS) is a very common sleep disorder. SAS is considered as clinically relevant when the breath stops during more than 10 seconds and occurs more than five times per sleep hour. In this paper, we present a noninvasive automatic approach to sleep apnea classification. Only noninvasive records of the respiratory and cardiac activities ( Nasal Airway Flow (NAF) and Pulse Transit Time (PTT)) issued by the technique of PolySomnoGraphy (PSG) are considered for the detection of the different sleep apnea syndromes: obstructive, central and hypopnea. Experimental results using clinical data are presented.

Polygonal Approximation Of Holter Registers: A Comparative Study For Electrographic Signals Time Compression

  • Autoři: Mico, P., Cuesta, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of the 2nd International conference on Computational Inteeligence in Medical and Healthcare, CIMED 2005. London: IEE, 2005, pp. 323-329. ISBN 0-86341-520-2.
  • Rok: 2005
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The development and improvement of biosignal recording devices implies a quality increase of the acquired signals that becomes an important problem on the treatment and storage of the big amount of information derived from this kind of records. We dealt with long-term Holter Electrocardiographic Signals (ECG) in order to apply analysis techniques (Hidden Markov Models, Principal Component Analysis, etc.) for processing the highest number of samples as fast as possible. For solving both, the storage-space and the time-processing problems, a comparative study among different approximation algorithms is presented, with the objective of reducing the size of the analysed signal and keeping samples containing a relevant piece of information. The results obtained from the experiments, induces us to select a simple metric to be used in the ECG compression process.

A Database of Occulographic Signals

  • Autoři: Cuesta Frau, D., doc. Ing. Daniel Novák, Ph.D., Aboy, M., Brzezny, R., Cerny, R., Jeřábek, J., Samblas-pena, L.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2004, pp. 97-99. ISSN 1211-412X. ISBN 80-214-2633-0.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We descibe a free, web-based occulographic signals database. The file forma tis chosen is the well-known and simple European Data Format (EDF). At the first stage, the database contains 39 registres with an IVIEW system, at the Motol Hospital, Prague, Czech Republic

A Novel Statistical Model for Simulation of Arterial and Intracranial Pressure

  • Autoři: Aboy, M., McNames, A., Hornero, R., Cuesta-Frau, D., Thong, T., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society Press, 2004, pp. 129-132. ISBN 0-7803-8440-7.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We describe a novel statistical model of pressure signals that incorporates the effects of respiration on arterial (ABP) and intracranial pressure (ICP). This model can be used to synthesize pulsatile ABP and ICP signals with similar time, frequency, and variability characteristics of real pressure signals. These synthetic signals can be used during the development, simulation, or quantitative assessment of biomedical algorithms in a variety of applications.

Clustering of Intracranial Pressure using Hidden Markov Models

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Lhotská, L., Cuesta-Frau, D., Abeo, M., Goldstein, B.
  • Publikace: Cybernetics and Systems 2004. Vienna: Austrian Society for Cybernetics Studies, 2004, pp. 175-180. ISBN 3-85206-169-5.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a clustering algorithm based on continuous Hidden Markov Models (HMM) to automatically classify intracranial pressure (ICP) beats based on their morphology. The algorithm detects, classifies and labels each beat as a low--pressure or high--pressure beat based on morphology. An ICP beat detection algorithm is used to automatically detect each beat. In order to avoid the numerical problems with classical Expectation-Maximization (EM) algorithm we applied Variational Bayes Learning for HMM optimization. We measured the performance of the algorithm compared to expert classification of ICP beats acquired from intensive care unit patients using both partitional and hierarchical clustering schemes. We showed that neither partitional nor hierarchical scheme are superior to each other; the clustering performance about 80 \% was achieved both on synthetic and real-icp data.

Complex Eye Movement Analysis and Its Use for Personal Computer Controlling

  • Autoři: Fejtová, M., Fejt, J., doc. Ing. Daniel Novák, Ph.D., Lhotská, L.
  • Publikace: Workshop 2004. Praha: České vysoké učení technické v Praze, 2004, pp. 810-811. ISBN 80-01-02945-X.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    In the project (working title MEMREC - Mobile Eye Movements RECorder) we are designing and developing a simple device for PC control.

Electroencephalogram processing using Hidden Markov Models

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Al-ani, T., Hamam, Y., Lhotská, L.
  • Publikace: Proceedings of the 5th EUROSIM Congres Modelling and Simulation. Vienna: EUROSIM-FRANCOSIM-ARGESIM, 2004, ISBN 3-901608-28-1.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    An approach for Electroencephalogram (EEG) processing is presented. Along with the theoretical development of stochastic processing techniques, two application areas are suggested: EEG sleep recording analysis and Brain Computer Interface (BCI). Many methods have been already developed in the area of sleep staging, nevertheless the automatic scoring in not still so effective as the manual scoring. Our sleep scoring method has the advantage of better temporal resolution (1 second) compared to the classical manual approach (30 seconds). In case of BCI this is a quite new approach offering mainly support for disable people in terms of controlling personal computer. The algorithm for cue movements determination has been designed resulting in detecting the movements within one second interval.

Feature Extraction From Biological Signals: A Case Study

  • Autoři: Lhotská, L., Fejtová, M., Macek, J., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: IEEE 4th International Conference on Intelligent Systems Design and Application. Piscataway: IEEE, 2004, pp. 139-144. ISBN 963-7154-29-9.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper presents two case studies illustrating the problem of data preprocessing as the first step in computer-aided analysis of biological signals used in clinical decision support. Methods for data extraction from ECG and EEG signals are described. We show the differences between these two signal types and the reasons why different approaches are used for their preprocessing. Analysis of ECG records is performed by the wavelet transform, and analysis of EEG records is performed using adaptive segmentation and the Fourier transform. The wavelet transform allows good localisation of QRS complexes, P and T waves in time and amplitude. The average accuracy of detection of all events is above 87 per cent. Adaptive segmentation abstracts the EEG signal data into stationary segments and the Fourier transform calculates their basic characteristics. In both cases extracted data are used as inputs for machine learning methods.

High-Speed Feature Extraction in Holter Electrocardiogram Using Principal Component Analysis

  • Autoři: Micó, P., Cuesta-Frau, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2004, pp. 81-83. ISSN 1211-412X. ISBN 80-214-2633-0.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Due to their long duration (up to 48 hours) and because of the enormous quantity of beats involved, it results really difficult to perform a manual inspection of HOLTER electrocardiographic signals (ECG). In the other side, we have patients with chronic heart-troubles whose heart rithm needs to be checked everytime using portable ECG recording machines. Within this scenario it would be very interesting the possibility of real-time ANALYZING and clustering ECG records in order to detect arrhythmia beats as soon as possible, allowing the patient to realize and prevent from a critical heart attack. To achieve this goal we have designed and implemented a C++ application that is based on the Principal Component Analysis (PCA) method applied to the beats (from a portable ECG recording machine) to decompose and cluster them in real-time. At the first stage of this article we will define holter ECG records and discuss about their desirable features for processing them ensuring better results y

Morphology Analysis of Physiological Signals Using Hidden Markov Models

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Micó, P., Al-ani, T., Hamam, Y., Aboy, M., Lhotská, L.
  • Publikace: 17th International Conference on Pattern Recognition. London: British Machine Vision Association, 2004, pp. 754-757. ISSN 1051-4651. ISBN 0-7695-2128-2.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We describe a clustering algorithm based on continuous Hidden Markov Models (HMM) to automatically classify both electrocardiogram (ECG) and intracranial pressure (ICP) beats based on their morphology. The algorithm detects, classifies and labels each beat based on morphology. In order to avoid the numerical problems with classical Expectation-Maximization (EM) algorithm we apply a novel method of simulated annealing (SIM) for HMM optimization. We show that better results are achieved using simulated annealing approach.

Pattern Matching Techniques Applied to Biomedical Signal Processing

  • Autoři: Cuesta-Frau, D., Micó-Tormos, P., doc. Ing. Daniel Novák, Ph.D., Aboy, M.
  • Publikace: International Journal of The International Institute for Advanced Studies in Systems Research and Cybernetics. 2004, IV(1), 29-35. ISSN 1609-8625.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Many biological processes produce outputs which can be characterized as signals and therefore, they can be easily registered with suitable medical equipment. This acquisition equipment is fitted to each specific case, there being many different registers utilized in clinical practice nowadays. These registers are also frequently processed and visualized with the same acquisition equipment, and physicians provide a diagnosis according to the information presented. Nevertheless, some of these registers need to be acquired over a long period, and therefore, raw visual inspection is very clumsy and difficult. Moreover, in recent years, many research groups have created web sites including biosignal databases to share with other groups worldwide. These databases usually contain hundreds of registers, and the amount of information is so enormous that it is almost impossible to examine all of them manually. For the cases mentioned above, it would be a breakthrough to find a non-supervised me.

Pre-clustering of Electrocardiographic Signals using Ergodic Hidden Markov Models

  • Autoři: Micó, P., Cuesa, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Structural, Syntactic, and Statistical Pattern Recognition. Berlin: Springer, 2004, pp. 939-947. ISSN 0302-9743. ISBN 3-540-22570-6.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Holter signals are ambulatory long-term electrocardiographic (ECG) registers used to detect heart diseases which are difficult to find in normal ECGs. These signals normally include several registers and its duration is up to 48 hours. The principal problem for the cardiologists consists of the manual inspection of the whole holter ECG to find all those beats whose morphology differ from the normal synus rhythm. The later analisys of these arrhythmia beats yields a diagnostic from the pacient's heart condition. Using Hidden Markov Models (HMM) for computer clustering has became a very useful tool for cardiologists avoiding the manual inspection. In this paper we improve the performance of the HMM clustering method introducing a preclustering stage in order to diminish the number of elements to be finally processed and reducing the global computational cost. An experimental comparative study is carried out, utilizing records form the MIT-BIH Arrhythmia database. Finally some results ar.

School Children Dyslexia Analysis using Self Organizing Maps

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Kordík, P., Macaš, M., Brzezny, R., Vyhnálek, M., Lhotská, L.
  • Publikace: 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society Press, 2004. p. 1-4. ISBN 0-7803-8440-7.
  • Rok: 2004
  • DOI: 10.1109/IEMBS.2004.1403075
  • Odkaz: https://doi.org/10.1109/IEMBS.2004.1403075
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The main goal of the study is an unsupervised classification of school children dyslexia. Eye movements of 49 subjects were measured using videooculographic technique (VOG) during two non-reading and one reading tasks. A feature selection was performed obtaining data set consisting of 26 features. Next an inductive modelling technique was applied to data set resulting in extraction of six features which were used as the input to self-organizing map (SOM). Three clusters were finally formed by the SOM proving that the proposed methodology is suitable for automatic dyslexia analysis.

Speech Recognition Methods Applied to Biomedical Signals Processing

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Mico-Tormos, P., Al-ani, T., Aboy, M., Lhotská, L.
  • Publikace: 26th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Los Alamitos: IEEE Computer Society Press, 2004, pp. 118-121. ISBN 0-7803-8440-7.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The paper focuses on processing of long biological signals used during monitoring procedures like in the case of portable Holter device for arrythmia analysis (ECG), intracranial pressure monitoring (ICP) in intensive care unit or overnight electroencephalogram monitoring (EEG) for sleep apnea detection. Two methods taken from speech processing are proposed: Dynamic Time Warping (DTW) and Hidden Markov Models (HMM). The unsupervised analysis of ECG and ICP beats is carried out using hierarchical clustering approach. In case of EEG, first the estimation of sleep stages is performed and next the different breathing events are detected by HMM by means of Viterbi inference. We show that for the first two problems DTW outperforms HMM while in the third case the HMM inference capability makes HMM suitable for sleep apnea diagnosis.

Unsupervised Learning of Holter ECG signals using HMM optimized by simulated annealing

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Al-ani, T., Hamam, Y., Cuesta Frau, D., Mico, P., Lhotská, L.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2004, pp. 60-62. ISSN 1211-412X. ISBN 80-214-2633-0.
  • Rok: 2004
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    We present a unsupervised learning algorithm based on continuous Hidden Markov Models (HMM) to automatically classify Holter signals based on their morphology. Our proposed method automatically detect and separate the significant beats by means of hierarchical clustering scheme. Due to the convergence and numeric problems of a classical local optimization technique, we have implemented a novel approach for the global training of HMM by simulated annealing

Biological Data Preprocessing: A Case Study

  • Autoři: Lhotská, L., Fejtová, M., Macek, J., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Intelligent and Adaptive Systems in Medicine. Praha: ČVUT v Praze, FEL, 2003, pp. 77-99. ISSN 1213-3000.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    This paper presents two case studies illustrating the problem of data pre-processing as the first step in computer-aided analysis of biological signals used in clinical decision support. Methods for data extraction from ECG and EEG signals are described. We show the differences between these two signal types and the reasons why different transforms are used for their pre-processing. Analysis of ECG records is performed by the wavelet transform, and analysis of EEG records is performed by the Fourier transform. The wavelet transform allows good localisation of QRS complexes, P and T waves in time and amplitude. The average accuracy of detection of all events is above 87 per cent. Adaptive segmentation abstracts the EEG signal data into stationary segments and the Fourier transform calculates their basic characteristics. In both cases extracted data are used as inputs for learning methods.

Biosignal Laboratory: A Software Tool for Biomedical Signal Processing and Analysis

  • Autoři: Cuesta-Frau, D., Mico Tormos, P., Aboy, M., doc. Ing. Daniel Novák, Ph.D., Brzezny, R.
  • Publikace: EMBC 2003. Piscataway: IEEE, 2003, pp. 3544-3547. ISSN 1094-687X. ISBN 0-7803-7790-7.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Biosignal processing and analysis is a field of great importance in current medical practice. In recent years, biomedical engineers have developed many algorithms and processing techniques in order to help doctors in the examination of many different biosignals, and to find new information embedded in them and not easily observable in the raw data. Since the number and variety of these techniques grows every day, it is becoming more and more difficult to be up to date in biosignal processing. Other aspects such as signals format, algorithms implementation and comparison, artifacts, etc, make the learning process in this field longer and longer, and in many occasions a lot of different software tools are needed to visualize or process biosignals. In this paper we present a program that consists of many of the current typical biosignal processing algorithms: filtering, domain transforms, basic operators, wave detection, and unlike other similar applications, it includes a pattern recogns

Independent Component Analysis and Its Applications

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Lhotská, L.
  • Publikace: Intelligent and Adaptive Systems in Medicine. Praha: ČVUT v Praze, FEL, 2003, pp. 100-108. ISSN 1213-3000.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Independent Component Analysis (ICA) is technique for separating mixtures of source signals into their individual components. This rapidly evolving technique is currently finding applications in speech separation, ECG, EEG, MEG or face recognition. Its power resides in the simple and realistic assumption that different physical process tend to generate statistically independent signals. In the first part of the survey we briefly provide mathematical background and in the second part we will explore the method usefulness on several real-life applications.

Neuropsychological screening of HIV encephalopathy

  • Autoři: Černý, R., Jung, M., Machala, L., Staňková, M., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Sixth IBRO World Congress of Neuroscience. Praha: AV ČR, 2003, ISBN 80-239-0887-1.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    HIV virus can invade central nervous system (CNS) early in the disease evolution cause a chronic subclinical encephalitis [1] . Clinical manifestation follows a asymptomatic period of variable length, when the immune system begins to be compromised and opportunistic infections, tumours or other complications tend to develop. Most important CNS affection, caused directly by the HIV virus, is the HIV encephalitis (HIVE). Prevalence of this complication was high in the begining of HIV pandemic, but declined substantially after introduction of zidovudin and highly active antiretroviral treatment into clinical practice [2], [3]. In more progressed immune deficit states when CD4 lymphocytes count drops to 200 or less per ml, prevalence of HIVE rises again to 10% - 20% of cases, with even more detected at autopsy [4, 5]. Incidence of clinically manifest HIVE is estimated at 7% in AIDS stage of HIV infection [6].

Number of arrhythmia beats determination in Holter electrocardiogram: How many clusters?

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Mico Tormos, P., Lhotská, L.
  • Publikace: EMBC 2003. Piscataway: IEEE, 2003, pp. 2845-2848. ISSN 1094-687X. ISBN 0-7803-7790-7.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    Holter signals correspond to long-term electrocardiograph (ECG) registers. Manual inspection of such signals is difficult because of the enormous quantity of beats involved. Throughout the literature several methods of automatically detecting and separating the significant beats using unsupervised learning were proposed. An important part of the unsupervised learning problem is determining the number of constituent clusters which best describe the data. In this paper we concentrate on the problem of the number of arrhythmia beats-clusters selection presented in Holter ECG. We apply and compare several criteria for assessing the number of clusters and we show that, with a Gaussian mixture model, the approach is able to select 'an optimal' number of arrhythmia beats and so partition a Holter ECG. The following criteria has been examined: Bayesian selection method, Akaike's information criteria, minimum description length, minimum message length, fuzzy hyper volume, evidence density and .

The Creation of Models of Biocybernetics Systems for Education

  • Autoři: Šorf, M., Eck, V., Janků, L., Fejtová, M., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Proceedings of Workshop 2003. Praha: České vysoké učení technické v Praze, 2003, pp. 870-871. ISBN 80-01-02708-2.
  • Rok: 2003
  • Pracoviště: Katedra kybernetiky
  • Anotace:
    The article describes biological data measurement and processing [1] and creation of models. The innovation is made in creation software for hearing signal processing, in creation methods for biological data processing [2], in creation of models of bio systém. noise level based on the finest scale coefficients. Being the signal divided in that way the denoising scheme is applied. Therefore, the foremost aim is the improvement of signal-to-noise ratio (SNR), not the minimisation of computational costs, since the denoising is carried out off-line

Automatic Extraction of Significant Beats from a Holter Register

  • Autoři: Cuesta-Frau, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002, pp. 3-5. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002

Clustering Electrocardiogrph Signals Using Hidden Markov Models

  • Autoři: Cuesta-Frau, D., doc. Ing. Daniel Novák, Ph.D., Micó-Tormos, P.
  • Publikace: IFMBE Proceedings. Wien: Technische Universität, 2002, pp. 362-363. ISSN 1680-0737. ISBN 3-901351-62-0.
  • Rok: 2002

Detection of Saccadic Eye Movements and Slow Phase Velocity Determination

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Brzezný, R., Černý, R., Eck, V.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2002, pp. 33-35. ISSN 1211-412X. ISBN 80-214-2120-7.
  • Rok: 2002

Early Dyslexia Detection Techniques by Means of Oculographic Signals

  • Autoři: Micó-Tormos, P., Cuesta-Frau, D., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: IFMBE Proceedings. Wien: Technische Universität, 2002, pp. 540-541. ISSN 1680-0737. ISBN 3-901351-62-0.
  • Rok: 2002

EEG Signal Proceesing Using Artificial Intelligence Methods

  • Autoři: Šorf, M., doc. Ing. Daniel Novák, Ph.D., Janků, L., Fejtová, M.
  • Publikace: Proceedings of Workshop 2002. Praha: České vysoké učení technické v Praze, 2002, pp. 900-901. ISBN 80-01-02511-X.
  • Rok: 2002

Feature Extraction Methods Applied to the Clustering of Electrocardiographic Signals. A Comparative Study

  • Autoři: Jiřina, M., doc. Ing. Daniel Novák, Ph.D., Pirez-Cortiz, J.C., Andreu-Garcia, G.
  • Publikace: ICPR 02: Proceedings 16th International Conference on Pattern Recognition. Los Alamitos: IEEE Computer Society Press, 2002, pp. 961-964. ISBN 0-7695-1695-X.
  • Rok: 2002

Method for Clinical Analysis of Eye Movements Induced by Rotation Test

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Cuesta-Frau, D., Brzezný, R., Černý, R., Lhotská, L.
  • Publikace: IFMBE Proceedings. Wien: Technische Universität, 2002, pp. 412-413. ISSN 1680-0737. ISBN 3-901351-62-0.
  • Rok: 2002

Application of the Method Machine Learning Approaches to Personality Type Classification

Applications of the Fuzzy Rule System and Neural Net to the Human Operator Stress Level Estimation

  • Autoři: Eck, V., Janků, L., Šorf, M., doc. Ing. Daniel Novák, Ph.D., Fejtová, M., Roknic, J.
  • Publikace: Proceedings. Zittau: Institut für Prozesstechnik und Messtechnik, 2001, pp. 191-194.
  • Rok: 2001

Denoising Electrocardiogram Signal Using Adaptive Wavelets

  • Autoři: doc. Ing. Daniel Novák, Ph.D., Eck, V., Cuesta Frau, D., Peréz-Cortés, J.C., Andreu-Garcia, G.
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2000, pp. 18-20. ISBN 80-214-1610-6.
  • Rok: 2000

Electrocardiogram Baseline Removal Using Wavelet Approximations

  • Autoři: Eck, V., Cuesta Frau, D., Peréz-Cortés, J.C., Andreu-Garcia, J., doc. Ing. Daniel Novák, Ph.D.,
  • Publikace: Analysis of Biomedical Signals and Images. Brno: VUTIUM Press, 2000, pp. 136-138. ISBN 80-214-1610-6.
  • Rok: 2000

Za stránku zodpovídá: Ing. Mgr. Radovan Suk