Parkinson’s disease affects more than 10 million people worldwide, and although there is still no cure, the disease can be better controlled if symptoms are caught early. One of the main aspects of Parkinson’s disease is that as it progresses, speech changes.
This led Rytis Maskeliūnas, a Lithuanian researcher from Kaunas University of Technology (KTU), and a team of colleagues from the Lithuanian University of Health Sciences (LSMU), to define and attempt to identify these early symptoms using voice data.
Detect subtle changes in speech patterns
According to Maskeliūnas, as motor activity decreases, the function of the vocal cords, diaphragm, and lungs also decreases.
“Changes in speech often occur even earlier than motor function disorders, which is why impaired speech could be the first sign of the disease,” explains Maskeliūnas.
Professor Virgilijus Ulozas from the Department of Ear, Nose and Throat at LSMU Medical School says patients with early-stage Parkinson’s disease might speak in a calmer, often monotonous, less expressive manner. , slower and more fragmented. That said, these changes are difficult to detect by ear.
The joint team of researchers has developed a new system that works to solve this problem.
“We are not creating a substitute for routine patient examination – our method is designed to aid in early diagnosis of disease and to monitor the effectiveness of treatment,” says Maskeliūnas.
He also says the link between the disease and speech abnormalities is not new, but as technology advances, it becomes easier to extract more insightful information from speech.
Use of AI algorithms and voice data
Using state-of-the-art artificial intelligence (AI), researchers conducted a groundbreaking study to create personalized analyzes and diagnostics of spoken signals in the Lithuanian language. Within seconds, they were able to extend existing AI databases with results unique to the linguistic peculiarities of Lithuania.
Kipras Pribuišis is a lecturer at the Department of Ear, Nose and Throat at the Faculty of Medicine at LSMU.
“So far, our approach is able to distinguish Parkinson’s disease from healthy people using a speech sample,” says Pribuišis. “This algorithm is also more accurate than the one previously proposed.”
By recording the speech of healthy patients and patients with Parkinson’s disease in a soundproof booth, the team used an AI algorithm to process the signals. This innovative approach required no complex hardware and could potentially be applied to mobile devices, paving the way for improved healthcare solutions in the future.
“Our results, which have already been published, have very high scientific potential. Of course, there is still a long and difficult way to go before it can be applied in daily clinical practice,” says Maskeliūnas.
Maskeliūnas identified the next steps in his research: increasing the number of patients to obtain additional evidence, comparing this algorithm with other methods of early detection of Parkinson’s disease and confirming its effectiveness in various contexts such as medical offices or home environments.