"Speech impairment is one of the first signs of Parkinson's disease. Together with colleagues, we have therefore tried to develop a mobile app that could detect problems by analysing phone calls," says Assoc. Petr Dušek, MUDr. Petr Dušek, Ph.D., from the Department of Neurology at the 1st Faculty of Medicine, Charles University in Prague, member of the National Institute for Neurological Research Neur-IN.
The experts involved over 70 participants in the study - in addition to people with early Parkinson's disease, they also monitored patients suffering from so-called REM sleep behaviour disorder. This is the first symptom of an emerging neurodegenerative disease. "We monitored two types of activity in the patients. If they gave consent, the app analysed their phone calls. It turned out that it only takes about 18 minutes to make a correct assessment, which corresponds to, for example, nine short calls with friends or family. At the same time, the participants had to read selected passages of text in the app, and the app reliably detected the fault after about 120 words," says Assoc. Dušek. This allowed the researchers to compare the differences in speech under different conditions - because the difficulties manifest themselves differently when a person speaks casually in front of you, and differently when they have to read out words in advance.
"In general, the voice of people with Parkinson's is more monotone - this distinguishes their speech from that of a healthy person. Speech melody, loudness and intelligibility are typically impaired. In the later stages of the disease, these difficulties are clearly audible. At the beginning, however, the changes may be very subtle - even these can be reliably detected by our app," the neurologist explains. Previously, his team has been able to describe three types of speech disorders that affect Parkinson's patients. "Depending on which disorder the patient suffers from, we are able to predict how their disease will develop and how they will respond to treatment. Patients with impaired articulation, who generally have a more severe course of the disease, respond the worst to therapy," says Assoc. Dušek.
Currently, the mobile app, developed at the Faculty of Electrical Engineering of the Czech Technical University, is in the testing phase - but experts are excited about the first results. "The application is very reliable. The results are comparable to analyses of calls we recorded with patients in laboratory conditions using a high-quality microphone. In the future, the mobile phone could really become a full-fledged tool to help us diagnose Parkinson's disease early on," says Assoc. Ing. Jan Rusz, Ph.D., from FEE CTU, in whose team the app was developed. According to the researchers, mass availability and ease of use are the main advantages of the developed software.
"Until now, speech analysis was carried out in such a way that the patient came to our clinic, where we put a professional microphone on him and he had to perform the assigned tasks in a quiet room for half an hour. Thanks to the mobile phone, everything has been simplified - now all a person has to do is download the app and give consent to send the calls for analysis," says Dr Vojtěch Illner from the FEE CTU, the lead author of the study.
Scientists presented the results of the groundbreaking study some time ago at the 2nd annual conference of the National Institute for Neurological Research, which took place in Prague. "The developed mobile application is absolutely unique. In the Czech Republic, we are still at the beginning of using similar technologies. Various researches are underway, for example, we are trying to develop an app for patients with functional neurological disorders at our department. However, the software for early diagnosis is exceptional - I consider its main advantage to be its user-friendliness," says Prof. Milan Brázdil, M.D., Ph.D., FRCP, Scientific Director of the National Institute for Neurological Research and Head of the First Neurological Clinic of the Faculty of Medicine of the National University of Medical Sciences and the Faculty of Medicine of the Medical University of Prague.
However, in order to translate the application into practice, according to doc. Dušek further testing is needed. The researchers would like to try other types of mobile devices and also speech analysis in other languages. "We are preparing for a time when we will have a treatment that could completely stop the progression of Parkinson's disease if the disease is detected early enough. Using passive biomarkers - such as our app - we could then easily screen across the population. People wouldn't have to undergo complicated tests, but would test themselves. And in the event of a positive finding, effective therapy would be deployed to stop the disease from developing," says doc. Dušek.
Parkinson's disease is a neurodegenerative disease of the central nervous system that is related to the loss of nerve cells. Patients gradually lose the ability to control their movements, develop tremors, and experience memory problems or other cognitive disorders. According to estimates, up to 50 000 people in the Czech Republic suffer from the disease. As a result of the ageing population, the number is increasing every year. It is not yet possible to cure the disease, but thanks to targeted therapy, doctors are able to suppress the symptoms and slow its development. As patients' condition progressively worsens, not only their quality of life declines, but also that of the families who care for them. However, if we can catch the disease at an early stage, we can extend the time that patients and their loved ones can lead a full life," adds Prof. Brázdil.
