Novosibirsk State Technical University has developed a voice screening system to assess the neurological status of patients with Parkinson's disease. The unique monitoring method has no analogues in Russia and is already being used at the Federal Center for Neurosurgery in Novosibirsk.
Maria Dunaeva, a fourth-year student of the Faculty of Automation and Computer Engineering at NSTU-NETI, proposed a method for determining the condition of patients with Parkinson's disease, a degenerative disease of the central nervous system associated with the death of dopamine-producing neurons in the substantia nigra of the brain.
One of the signs of the disease is speech problems and a quiet monotonous voice, reflecting the severity of the patient's condition. It is suggested to evaluate the neurological status and monitor the condition of patients with Parkinson's disease using a special web application. The development allows you to analyze the main characteristics of the voice signal (changes in the voice in Parkinson's disease by special calculations of the average amplitude, duration, pauses, tremors and interruptions of the voice), lexical meanings of speech (the number of complex sentences, lexical diversity, frequency of use of different parts of speech) and provides a more detailed medical history.
According to Maria Dunaeva, the application consists of three modules: voice screening, medical history management and reporting. Together, they allow you to analyze the characteristics of recorded voice samples, store patient information, including data on visits and prescribed medications, and create reports based on the analysis for diagnosis. The application is relevant for all stages of the disease, including severe ones, but most importantly, it is aimed at sharing with doctors.
"The development is aimed at joint use with a doctor, including in the postoperative period. There is data synchronization between the doctor and the patient, thanks to which the doctor can quickly track the dynamics of the disease. Self—diagnosis can only harm the patient — there is a risk that he will come up with a diagnosis for himself," said Maria Dunaeva.
In the future, it is planned to transform the system into a module format for medical information systems. "In practice, it looks like this: during postoperative rehabilitation, a patient opens a website, logs into his personal account and sees prescribed medications, examination results and voice screening results. He is asked to undergo weekly screening to track his performance. Then, after a while, the doctor sees all the previous results at the reception and can conclude how much the indicators have improved," explained Maria.
The development of the NSTU-NETI student was tested and highly appreciated by doctors at the XVI Russian Final Scientific and Practical Conference with international participation of students and young scientists "Avicenna-2025". According to Maria Dunaeva, the doctors noted the simplicity of calculating the quantitative characteristics of the voice and the ability to track the dynamics in the postoperative period, including at home, when the patient records the voice himself, and doctors analyze it.
The voice screening system for assessing the neurological status of patients with Parkinson's disease has already been implemented in the Federal State Budgetary Institution Federal Center for Neurosurgery of the Ministry of Health of the Russian Federation (Novosibirsk). According to the staff of the center, a non-invasive method for diagnosing the disease based on the analysis of voice changes due to neurodegenerative processes in the brain allows the attending physician to monitor changes in patients and record both an improvement in therapy and a deterioration in the condition.
"The advantage of the system is the possibility of early diagnosis of violations. In addition, a big advantage is the ability to track changes in the patient's condition from a distance. In the future, with the help of this development, we plan to diagnose voice changes for other neurological diseases, which will allow us to find certain features of the voice for differential diagnosis between pathologies," said Galina Moisak, a neurologist at the Federal Research Center, Candidate of Medical Sciences.