bsuirДоклады БГУИРDoklady BGUIR1729-76482708-0382БГУИР10.35596/1729-7648-2025-23-1-68-73bsuir-4065Research ArticleСтатьиМашинное обучение и нейронные сети для IT-диагностики неврологических заболеванийMachine Learning and Neural Networks for IT-Diagnostics of Neurological DiseasesВишняковВ. А.VishniakouU. A.

Вишняков Владимир Анатольевич, д-р техн. наук, проф каф. инфокоммуникационных технологий

220013, г. Минск, ул. П. Бровки, 6 

Vishniakou Uladzimir Anatol’evich, Dr. of Sci. (Tech.), Professor at the Department of Infocommunication Technologies

220013, Minsk, P. Brovki St., 6 

vish@bsuir.byСяИ. В.XiaY. W.

асп. каф. инфокоммуникационных технологий

г. Минск

Postgraduate at the Department of Infocommunication Technologies

Minsk

ЮйЧ. Ю.YuCh. Y.

асп. каф. инфокоммуникационных технологий

г. Минск

Postgraduate at the Department of Infocommunication Technologies

Minsk

Белорусский государственный университет информатики и радиоэлектроникиBelarusian State University of Informatics and Radioelectronics2025170220252316873Copyright © Вишняков В.А., Ся И.В., Юй Ч.Ю., 20252025Вишняков В.А., Ся И.В., Юй Ч.Ю.Vishniakou U.A., Xia Y.W., Yu C.Y.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://doklady.bsuir.by/jour/article/view/4065

Рассмотрены методы машинного обучения и нейронные сети для диагностики неврологических заболеваний (болезней Альцгеймера и Паркинсона) пациентов на основе голосового анализа. Приведены модели извлеченной из голосовых данных информации о признаках заболеваний (включая частоту, дрожание, мел-кепстральные коэффициенты и т. д.). Использованы различные классификаторы для обучения нейронных сетей и распознавания заболеваний. Среди них – алгоритм GridSearchCV для оптимизации гиперпараметров классификатора случайного леса при распознавании болезни Альцгеймера (точность распознавания – 87,6 %) и алгоритм KNN – для обучения и тестирования на общедоступных наборах данных признаков изменения речи пациентов с болезнью Паркинсона. Алгоритм KNN показал лучшие результаты классификации по сравнению с другими, достигнув экспериментальной точности 94 % на тех же наборах данных. Отмечено, что использование многомерного извлечения признаков и методов машинного обучения может повысить точность ранней диагностики неврологических заболеваний.

The article considers machine learning methods and neural networks for diagnosing neurological diseases (Alzheimer’s and Parkinson’s diseases) in patients based on voice analysis. Models of information about disease features (including frequency, jitter, mel-cepstral coefficients, etc.) extracted from voice data are presented. Various classifiers are used to train neural networks and recognize diseases. Among them are the GridSearchCV algorithm for optimizing the hyperparameters of the random forest classifier for recognizing Alzheimer’s disease (recognition accuracy is 87.6 %) and the KNN algorithm for training and testing on publicly available datasets of speech change features in patients with Parkinson’s disease. The KNN algorithm showed the best classification results compared to others, achieving an experimental accuracy of 94 % on the same datasets. It is noted that the use of multidimensional feature extraction and machine learning methods can improve the accuracy of early diagnosis of neurological diseases. 

диагностика пациентовизменение голосаневрологические заболеваниямашинное обучениенейронные сетиpatient diagnosticsvoice changeneurological diseasesmachine learningneural networksReferencesRaghavendra, U. Artificial Intelligence Techniques for Automated Diagnosis of Neurological Disorders / U. Raghavendra, U. R. Acharya, H. Adeli // European Neurology. 2019. Vol. 82, No 1–3. Р. 41–64.Raghavendra U., Acharya U. R., Adeli H. (2019) Artificial Intelligence Techniques for Automated Diagnosis of Neurological Disorders. European Neurology. 82 (1–3), 41–64.Armstrong, M. J. Diagnosis and TReatment of Parkinson Disease: A Review / M. J. Armstrong, S. O. Michael // Jama. 2020. Vol. 323, No 6. Р. 548–560.Armstrong M. J., Michael S. O. (2020) Diagnosis and Treatment of Parkinson Disease: A Review. Jama. 323 (6), 548–560.Early Diagnosis of Parkinson’s Disease / G. Becker [et al.] // Journal of Neurology. 2002. Vol. 249. Р. iii40–iii48.Becker G., Müller A., Braune S., Büttner T., Benecke R., Greulich W., et al. (2002) Early Diagnosis of Parkinson’s Disease. Journal of Neurology. 249, iii40–iii48.Agarwal, A. Prediction of Parkinson’s Disease Using Speech Signal with Extreme Learning Machine / A. Agarwal, S. Chandrayan, S. S. Sahu // 2016 International Conference on Electrical, Electronics, and Optimization Techniques. Р. 3776–3779.Agarwal A., Chandrayan S., Sahu S. S. (2016) Prediction of Parkinson’s Disease Using Speech Signal with Extreme Learning Machine. 2016 International Conference on Electrical, Electronics, and Optimization Techniques. 3776–3779.Blahuta, J. An Expert System Based on Using Artificial Neural Network and Region-Based Image Processing to Recognition Substantia Nigra and Atherosclerotic Plaques in B-Images: A Prospective Study / J. Blahuta, T. Soukup, J. Martinu // Advances in Computational Intelligence, 14th International Work-Conference on Artificial Neural Networks, Cadiz, Spain, June 14–16, 2017, Part I. Cadiz, Spain: Springer International Publ. Р. 236–245.Blahuta J., Soukup T., Martinu J. (2017) An Expert System Based on Using Artificial Neural Network and Region-Based Image Processing to Recognition Substantia Nigra and Atherosclerotic Plaques in B-Images: A Prospective Study. Advances in Computational Intelligence, 14th International Work-Conference on Artificial Neural Networks, Cadiz, Spain, June 14–16, Part I. Springer International Publ. 236–245.Convolution Neural Networks and Self-Attention Learners for Alzheimer Dementia Diagnosis from Brain MRI / P. Carcagnì [et al.] // Sensors. 2023. Vol. 23, No 3.Carcagnì P., Leo M., Del Coco M., Distante C., De Salve A. (2023) Convolution Neural Networks and Self-Attention Learners for Alzheimer Dementia Diagnosis from brain MRI. Sensors. 23 (3).SpeechFormer++: A Hierarchical Efficient Framework for Paralinguistic Speech Processing / W. Chen [et al.] // arXiv:2302.14638v1. 2023. Vol. 31. Р. 775–788.Chen W., Xing X., Xu X., Pang J., Du L. (2023) SpeechFormer++: A Hierarchical Efficient Framework for Paralinguistic Speech Processing. arXiv:2302.14638v1. 31, 775–788.Using the Outputs of Different Automatic Speech Recognition Paradigms for Acoustic- and BERT-Based Alzheimer’s Dementia Detection Through Spontaneous Speech / Y. Pan [et al.] // Interspeech. 2021. Р. 3810–3814.Pan Y., Mirheidari B., Harris J. M., Thompson J. C., Jones M., Snowden J. S., et al. (2021) Using the Outputs of Different Automatic Speech Recognition Paradigms for Acoustic- and BERT-Based Alzheimer’s Dementia Detection Through Spontaneous Speech. Interspeech. 3810–3814.A Scoping Review of Mathematical Models Covering Alzheimer’s Disease Progression / S. Moravveji [et al.] // Frontiers in Neuroinformatics. 2024. Vol. 18.Moravveji S., Doyon N., Mashreghi J., Duchesne S. (2024) A Scoping Review of Mathematical Models Covering Alzheimer’s Disease Progression. Frontiers in Neuroinformatics. 18.Вишняков, Ю. А. Использование машинного обучения для распознавания болезни Альцгеймера на основе транскрипционной информации / Ю. А. Вишняков, Юй Чу Юэ // Доклады БГУИР. 2023. Т. 21, № 6. С. 106–112. http://dx.doi.org/10.35596/1729-7648-2023-21-6-106-112.Vishniakou U. A., Yu Chu Yue (2023) Using Machine Learning for Recognition of Alzheimer’s Disease Based on Transcription Information. Doklady BGUIR. 21 (6), 106–112. http://dx.doi.org/10.35596/1729-7648-202321-6-106-112.Vishniakou, U. A. Technology of Neurological Disease Recognition Using Gated Recurrent Unite Neural Network and Internet of Things / U. A. Vishniakou, YiWei Xia, Chuyue Yu // OSTES Research Papers Collection. Minsk: Belarusian State University of Informatics and Radioelectronics, 2023. Р. 241–246.Vishniakou U. A., YiWei Xia, Chuyue Yu (2023) Technology of Neurological Disease Recognition Using Gated Recurrent Unite Neural Network and Internet of Things. OSTES Research Papers Collection. Minsk, Belarusian State University of Informatics and Radioelectronics. 241–246.Alzheimer’s Dementia Recognition Through Spontaneous Speech: The ADReSS Challenge / S. Luz [et al.] // arXiv:2004.06833. 2004.Luz S., Haider F., de la Fuente S., Fromm D., MacWhinney B. (2004) Alzheimer’s Dementia Recognition Through Spontaneous Speech: The ADReSS Challenge. arXiv:2004.06833.Haulcy, R. Classifying Alzheimer’s Disease Using Audio and Text-Based Representations of Speech / R. Haulcy, J. Glass // Frontiers in Psychology. 2021. Vol. 11.Haulcy R., Glass J. (2021) Classifying Alzheimer’s Disease Using Audio and Text-Based Representations of Speech. Frontiers in Psychology. 11.A Comparative Analysis of Speech Signal Processing Algorithms for Parkinson’s Disease Classification and the Use of the Tunable Q-factor Wavelet Transform / C. O. Sakar [et al.] // Applied Soft Computing. 2019. Vol. 74. Р. 255–263.Sakar C. O., Serbes G., Gunduz A., Tunc H. C., Nizam H., Sakar B. E., et al. (2019) A Comparative Analysis of Speech Signal Processing Algorithms for Parkinson’s Disease Classification and the Use of the Tunable Q-factor Wavelet Transform. Applied Soft Computing. 74, 255–263.Вишняков, В. А. IТ-диагностика болезни Паркинсона на основе анализа голосовых маркеров и машинного обучения / В. А. Вишняков, Ивэй Ся // Доклады БГУИР. 2023. Т. 21, № 3. С. 102–110.http://dx.doi.org/10.35596/1729-7648-2023-21-3-102-110.Vishniakou U. A., YiWei Xia (2023) IT Diagnostics of Parkinson’s Disease Based on the Analysis of Voice Markers and Machine Learning. Doklady BGUIR. 21 (3), 102–110. http://dx.doi.org/10.35596/1729-76482023-21-3-102-110.

The authors declare that there are no conflicts of interest present.