Ing. Martin Šubert, Ph.D.

Background and objectives Patients with synucleinopathies such as multiple system atrophy (MSA) and Parkinson’s disease (PD) frequently display speech and language abnormalities. We explore the diagnostic potential of automated linguistic analysis of natural spontaneous speech to differentiate MSA and PD. Methods Spontaneous speech of 39 participants with MSA compared to 39 drug-naive PD and 39 healthy controls matched for age and sex was transcribed and linguistically annotated using automatic speech recognition and natural language processing. A quantitative analysis was performed using 6 lexical and syntactic and 2 acoustic features. Results were compared with human-controlled analysis to assess the robustness of the approach. Diagnostic accuracy was evaluated using sensitivity analysis. Results Despite similar disease duration, linguistic abnormalities were generally more severe in MSA than in PD, leading to high diagnostic accuracy with an area under the curve of 0.81. Compared to controls, MSA showed decreased grammatical component usage, more repetitive phrases, shorter sentences, reduced sentence development, slower articulation rate, and increased duration of pauses, whereas PD had only shorter sentences, reduced sentence development, and longer pauses. Only slower articulation rate was distinctive for MSA while unchanged for PD relative to controls. The highest correlation was found between bulbar/pseudobulbar clinical score and sentence length (r = −0.49, p = 0.002). Despite the relatively high severity of dysarthria in MSA, a strong agreement between manually and automatically computed results was achieved. Discussion Automated linguistic analysis may offer an objective, cost-effective, and widely applicable biomarker to differentiate synucleinopathies with similar clinical manifestations.

Background and objectivesThe impact of dopaminergic medication on language in Parkinson's disease (PD) remains poorly understood. This observational, naturalistic study aimed to investigate the effects of long-term dopaminergic therapy on language performance in patients with de-novo PD based on a high-level linguistic analysis of natural spontaneous discourse.MethodsA fairy-tale narration was recorded at baseline and a 12-month follow-up. The speech samples were automatically analyzed using six representative lexical and syntactic features based on automatic speech recognition and natural language processing.ResultsWe enrolled 109 de-novo PD patients compared to 68 healthy controls. All subjects completed the 12-month follow-up; 92 PD patients were on stable dopaminergic medication (PD-treated), while 17 PD patients remained without medication (PD-untreated). At baseline, the PD-treated group exhibited abnormalities in syntactic domains, particularly in sentence length (p = 0.018) and sentence development (p = 0.042) compared to healthy controls. After 12 months of dopaminergic therapy, PD-treated showed improvements in the syntactic domain, including sentence length (p = 0.012) and sentence development (p = 0.030). Of all PD-treated patients, 37 were on monotherapy with dopamine agonists and manifested improvement in sentence length (p = 0.048), while 32 were on monotherapy with levodopa and had no language amelioration. No changes in language parameters over time were seen in both the PD-untreated group and healthy controls.DiscussionInitiation of dopaminergic therapy improved high-language syntactic deficits in de-novo PD, confirming the role of dopamine in cognitive-linguistic processing. Automated linguistic analysis of spontaneous speech via natural language processing can assist in improving the prediction and management of language deficits in PD.

Background: Language dysfunction is common in multiple sclerosis (MS). Speech analysis utilizing automated techniques offers potential for detecting subtle cognitive changes through quantification of language features derived from spontaneous speech, an easily acquired, naturalistic measure that may be a proxy of general cognitive function. Objectives: To determine associations of language features to objective and subjective cognition (primary), mood and fatigue (secondary), and evaluate discriminability. Methods: Participants with MS and controls completed objective and subjective cognitive measures and provided a 2-min spontaneous speech sample that was analyzed using natural language processing. Convergent validity was assessed by calculating correlations of derived language features to objective and subjective cognition. Correlations of language features to fatigue, anxiety, and depression were also evaluated. Finally, validity of language features to discriminate individuals with MS from controls was assessed. Results: In the full sample, three language features- N-grams, content words, and function words- were correlated with objective cognitive test performance (r's from 0.3 to 0.4). In the MS group, language features were associated with subjective cognition (r=-0.361, p=.007) and three objective tests (r's from 0.3 to 0.5), and showed no association to fatigue, anxiety, or depression. Compared to controls, the MS group produced a lower number of adverbs (p=.001). Conclusion: Semi-automated speech analysis to derive language variables may provide early indications of cognitive changes in MS, potentially enhancing diagnosis and patient management strategies.

Objective: This study assessed the relationship between speech and language impairment and outcome in a multicenter cohort of isolated/idiopathic rapid eye movement (REM) sleep behavior disorder (iRBD).Methods: Patients with iRBD from 7 centers speaking Czech, English, German, French, and Italian languages underwent a detailed speech assessment at baseline. Story-tale narratives were transcribed and linguistically annotated using fully automated methods based on automatic speech recognition and natural language processing algorithms, leading to the 3 distinctive linguistic and 2 acoustic patterns of language deterioration and associated composite indexes of their overall severity. Patients were then prospectively followed and received assessments for parkinsonism or dementia during follow-up. The Cox proportional hazard was performed to evaluate the predictive value of language patterns for phenoconversion over a follow-up period of 5 years.Results: Of 180 patients free of parkinsonism or dementia, 156 provided follow-up information. After a mean follow-up of 2.7 years, 42 (26.9%) patients developed neurodegenerative disease. Patients with higher severity of linguistic abnormalities (hazard ratio [HR = 2.35]) and acoustic abnormalities (HR = 1.92) were more likely to develop a defined neurodegenerative disease, with converters having lower content richness (HR = 1.74), slower articulation rate (HR = 1.58), and prolonged pauses (HR = 1.46). Dementia-first (n = 16) and parkinsonism-first with mild cognitive impairment (n = 9) converters had higher severity of linguistic abnormalities than parkinsonism-first with normal cognition converters (n = 17).Interpretation: Automated language analysis might provide a predictor of phenoconversion from iRBD into synucleinopathy subtypes with cognitive impairment, and thus can be used to stratify patients for neuroprotective trials. ANN NEUROL 2023

Background:Impairment of higher language functions associated with natural spontaneous speech in multiple sclerosis (MS) remains underexplored. Objectives:We presented a fully automated method for discriminating MS patients from healthy controls based on lexical and syntactic linguistic features. Methods:We enrolled 120 MS individuals with Expanded Disability Status Scale ranging from 1 to 6.5 and 120 age-, sex-, and education-matched healthy controls. Linguistic analysis was performed with fully automated methods based on automatic speech recognition and natural language processing techniques using eight lexical and syntactic features acquired from the spontaneous discourse. Fully automated annotations were compared with human annotations. Results:Compared with healthy controls, lexical impairment in MS consisted of an increase in content words (p = 0.037), a decrease in function words (p = 0.007), and overuse of verbs at the expense of noun (p = 0.047), while syntactic impairment manifested as shorter utterance length (p = 0.002), and low number of coordinate clause (p < 0.001). A fully automated language analysis approach enabled discrimination between MS and controls with an area under the curve of 0.70. A significant relationship was detected between shorter utterance length and lower symbol digit modalities test score (r = 0.25, p = 0.008). Strong associations between a majority of automatically and manually computed features were observed (r > 0.88, p < 0.001). Conclusion:Automated discourse analysis has the potential to provide an easy-to-implement and low-cost language-based biomarker of cognitive decline in MS for future clinical trials.

Background: Patients with synucleinopathies frequently display language abnormalities. However, whether patients with isolated rapid eye movement sleep behavior disorder (iRBD) have prodromal language impairment remains unknown. Objectives: We examined whether the linguistic abnormalities in iRBD can serve as potential biomarkers for conversion to synucleinopathy, including the possible effect of mild cognitive impairment (MCI), speaking task, and automation of analysis procedure. Methods: We enrolled 139 Czech native participants, including 40 iRBD without MCI and 14 iRBD with MCI compared to 40 PD without MCI, 15 PD with MCI, and 30 healthy controls. Spontaneous discourse and story tale narrative were transcribed and linguistically annotated. A quantitative analysis was performed computing 3 linguistic features. Human annotations were compared to fully-automated annotations. Results: Compared to controls, iRBD patients showed poorer content density, reflecting the reduction of content words and modifiers. Both PD and iRBD subgroups with MCI manifested less occurrence of unique words and a higher number of n-grams repetitions, indicating poorer lexical richness. The spontaneous discourse task demonstrated language impairment in iRBD without MCI with an area under the curve of 0.72, while the story tale narrative task better reflected the presence of MCI, discriminating both PD and iRBD subgroups with MCI from controls with an area under the curve of up to 0.81. A strong correlation between manually and automatically computed results was achieved. Conclusions: Linguistic features might provide a reliable automated method for detecting cognitive decline due to prodromal neurodegeneration in subjects with iRBD, providing critical outcomes for future therapeutic trials.