informИнформатикаInformatics1816-03012617-6963UIIP NASB10.37661/1816-0301-2023-20-3-106-114inform-1249Research ArticleИНТЕЛЛЕКТУАЛЬНЫЕ СИСТЕМЫINTELLIGENT SYSTEMSРаспознавание признаков болезни Паркинсона на основе анализа голосовых маркеров и двигательной активностиRecognition of signs of Parkinson's disease based on the analysis of voice markers and motor activityhttps://orcid.org/0000-0003-2929-8958ВишняковВ. А.VishniakouU. A.

Вишняков Владимир Анатольевич - доктор технических наук, профессор кафедры инфокоммуникационных технологий.

ул. П. Бровки, 6, Минск, 220013

Uladzimir A. Vishniakou - D. Sc. (Eng.), Professor of the Department of Infocommunication Technologies, Belarusian State University of Informatics and Radioelectronics.

Brovka P. st., 6, Minsk, 220013

vish@bsuir.byИвэйСяYiweiXia

Ся Ивэй - аспирант кафедры инфокоммуникационных технологий.

ул. П. Бровки, 6, Минск, 220013

Xia Yiwei - Postgraduate Student of the Department of Infocommunication Technologies, Belarusian State University of Informatics and Radioelectronics.

Brovka P. st., 6, Minsk, 220013

vish@bsuir.byБелорусский государственный университет информатики и радиоэлектроникиBelarusian State University of Informatics and Radioelectronics202329092023203106114Copyright © Вишняков В.А., Ивэй С., 20232023Вишняков В.А., Ивэй С.Vishniakou U.A., Yiwei X.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://inf.grid.by/jour/article/view/1249Цели

Цели. Решается задача ИТ-диагностики признаков болезни Паркинсона по анализу изменения голоса и замедления движения пациентов. Актуальность задачи связана с необходимостью ранней диагностики заболевания. Предлагается метод комплексного распознавания болезни Паркинсона с использованием машинного обучения, основанный на анализе голосовых маркеров и изменений в движениях пациентов на известных наборах данных.

Методы

Методы. Используются частотно-временная функция (функция вейвлета), функция кепстрального коэффициента Мейера, алгоритм k-ближайших соседей (k-Nearest Neighbors, KNN), алгоритм двухслойной нейронной сети для обучения и тестирования на общедоступных наборах данных по изменению речи и замедлению движения при болезни Паркинсона, а также байесовский оптимизатор для улучшения гиперпараметров алгоритма KNN.

Результаты

Результаты. Алгоритм KNN использован для распознавания речи пациентов, точность теста 94,7 % достигнута при диагностике болезни Паркинсона по изменению голоса. Алгоритм байесовской нейронной сети применен для распознавания замедления движения пациентов, он дал точность теста 96,2 %.

Заключение

Заключение. Полученные результаты распознавания признаков болезни Паркинсона близки к мировому уровню. На том же наборе данных по изменению речи пациентов один из лучших показателей зарубежных исследований составляет 95,8 %, а на наборе данных по замедлению движения пациентов - 98,8 %. Предлагаемая авторская методика предназначена для использования в подсистеме ИТ-диагностики неврологических заболеваний умного города.

Objectives

Objectives. The problem of IT diagnostics of signs of Parkinson's disease is solved by analyzing changes in the voice and slowing down the movement of patients. The urgency of the task is associated with the need for early diagnosis of the disease. A method of complex recognition of Parkinson's disease using machine learning is proposed, based on markers of voice analysis and changes in the patient's movements on known data sets.

Methods

Methods. The time-frequency function (the wavelet function) and the Meyer kepstral coefficient function, the KNN algorithm (k-Nearest Neighbors, KNN) and the algorithm of a two-layer neural network are used for training and testing on publicly available datasets on speech changes and motion retardation in Parkinson's disease. A Bayesian optimizer is also used to improve the hyperparameters of the KNN algorithm.

Results

Results. The KNN algorithm was used for speech recognition of patients, the test accuracy of 94.7% was achieved in the diagnosis of Parkinson's disease by voice change. The Bayesian neural network algorithm was applied to recognize the slowing down of the patients' movements, it gave a test accuracy of 96.2% for the diagnosis of Parkinson's disease.

Conclusion

Conclusion. The obtained results of recognition of signs of Parkinson's disease are close to the world level. On the same set of data on speech changes of patients, one of the best indicators of foreign studies is 95.8%. On the same set of data on motion deceleration, one of the best indicators of foreign researchers is 98.8%. The proposed author's technique is intended for use in the subsystem of IT diagnostics of neurological diseases of a Smart city.

распознавание болезни Паркинсонамашинное обучениеалгоритм KNNголосовые маркерыбайесовская нейронная сетьзамедление движениягиперпараметрыParkinson's disease recognitionmachine learningKNN algorithmvoice markersBayesian neural networkmotion decelerationhyperparametersReferencesDavie C. A. A review of Parkinson's disease. British Medical Bulletin, Feb. 2008, vol. 86, no. 1, pp. 109-127. https://doi.org/10.1093/bmb/ldn013Davie C. A. A review of Parkinson's disease. British Medical Bulletin, Feb. 2008, vol. 86, no. 1, pp. 109-127. https://doi.org/10.1093/bmb/ldn013Braak H., Ghebremedhin E., Rub U., Bratzke H., Tredici K. D. Stages in the development of Parkinson's disease-related pathology. Cell and Tissue Research, October 2004, vol. 318, no. 1, pp. 121-134. https://doi.org/10.1007/s00441-004-0956-9Braak H., Ghebremedhin E., Rub U., Bratzke H., Tredici K. D. Stages in the development of Parkinson's disease-related pathology. Cell and Tissue Research, October 2004, vol. 318, no. 1, pp. 121-134. https://doi.org/10.1007/s00441-004-0956-9Upadhya S. S., Cheeran A. N. Discriminating Parkinson and healthy people using phonation and cepstral features of speech. Procedia Computer Science, January 2018, vol. 143, pp. 197-202. https://doi.org/10.1016/j.procs.2018.10.376Upadhya S. S., Cheeran A. N. Discriminating Parkinson and healthy people using phonation and cepstral features of speech. Procedia Computer Science, January 2018, vol. 143, pp. 197-202. https://doi.org/10.1016/j.procs.2018.10.376Voiceprint recognition of Parkinson Patients Based on Deep Learning [Electronic resource] / Z. Xu [et al.]. - Dec. 2018. - Р. 1-10. - Mode of access: https://arxiv.org/ftp/arxiv/papers/1812/1812.06613.pdf. - Date of access: 12.11.2022.Voiceprint recognition of Parkinson Patients Based on Deep Learning [Electronic resource] / Z. Xu [et al.]. - Dec. 2018. - Р. 1-10. - Mode of access: https://arxiv.org/ftp/arxiv/papers/1812/1812.06613.pdf. - Date of access: 12.11.2022.Transformers for 1D Signals in Parkinson's Disease Detection from Gait [Electronic resource] / D. M. D. Nguyen [et al.]. - Apr. 2022. - Р. 1-7. - Mode of access: https://arxiv.org/pdf/2204.00423.pdf. - Date of access: 12.11.2022.Transformers for 1D Signals in Parkinson's Disease Detection from Gait [Electronic resource] / D. M. D. Nguyen [et al.]. - Apr. 2022. - Р. 1-7. - Mode of access: https://arxiv.org/pdf/2204.00423.pdf. - Date of access: 12.11.2022.Arshad H., Khan M. A., Sharif M., Yasmin M., Javed M. Y. Multi-level features fusion and selection for human gait recognition: an optimized framework of Bayesian model and binomial distribution. International Journal of Machine Learning and Cybernetics, December 2018, vol. 10, no. 12, pp. 3601-3618. https://doi.org/10.1007/s13042-019-00947-0Arshad H., Khan M. A., Sharif M., Yasmin M., Javed M. Y. Multi-level features fusion and selection for human gait recognition: an optimized framework of Bayesian model and binomial distribution. International Journal of Machine Learning and Cybernetics, December 2018, vol. 10, no. 12, pp. 3601-3618. https://doi.org/10.1007/s13042-019-00947-0Maachi I. E., Bilodeau G.-A., Bouachir W. Deep 1D-Convnet for accurate Parkinson disease detection and severity prediction from gait. Expert Systems with Applications, May 2020, vol. 143, pp. 1-27. https://doi.org/10.1016/j.eswa.2019.113075Maachi I. E., Bilodeau G.-A., Bouachir W. Deep 1D-Convnet for accurate Parkinson disease detection and severity prediction from gait. Expert Systems with Applications, May 2020, vol. 143, pp. 1-27. https://doi.org/10.1016/j.eswa.2019.113075Moro-Velazquez L., Gomez-Garcia J. A., Arias-Londoño J. D., Dehak N., Godino-Llorente J. I. Advances in Parkinson's Disease detection and assessment using voice and speech: A review of the articulatory and phonatory aspects. Biomedical Signal Processing and Control, April 2021, vol. 66, pp. 1-13. https://doi.org/10.1016/j.bspc.2021.102418Moro-Velazquez L., Gomez-Garcia J. A., Arias-Londoño J. D., Dehak N., Godino-Llorente J. I. Advances in Parkinson's Disease detection and assessment using voice and speech: A review of the articulatory and phonatory aspects. Biomedical Signal Processing and Control, April 2021, vol. 66, pp. 1-13. https://doi.org/10.1016/j.bspc.2021.102418Zhang M.-L., Zhou Z.-H. ML-KNN: A lazy learning approach to multi-label learning. Pattern Recognition, 2007, vol. 40, no. 7, pp. 1-21.Zhang M.-L., Zhou Z.-H. ML-KNN: A lazy learning approach to multi-label learning. Pattern Recognition, 2007, vol. 40, no. 7, pp. 1-21.Sakar C. O., Serbes G., Gunduz A., Tunc H. C., Nizam H., ..., Apaydin H. 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, January 2019, vol. 74, pp. 255-263. https://doi.org/10.1016/j.asoc.2018.10.022Sakar C. O., Serbes G., Gunduz A., Tunc H. C., Nizam H., ..., Apaydin H. 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, January 2019, vol. 74, pp. 255-263. https://doi.org/10.1016/j.asoc.2018.10.022Sakar B. E., Isenkul M. E., Sakar C. O., A. Sertbas, Gurgen F., Kursun O. Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings. IEEE Journal of Biomedical and Health Informatics, 2013, vol. 17(4), pp. 828-834.Sakar B. E., Isenkul M. E., Sakar C. O., A. Sertbas, Gurgen F., Kursun O. Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings. IEEE Journal of Biomedical and Health Informatics, 2013, vol. 17(4), pp. 828-834.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861-874.Fawcett T. An introduction to ROC analysis. Pattern Recognition Letters, 2006, vol. 27, no. 8, pp. 861-874.Li B., Yao Z., Wang J., Wang S., Yang X., Sun Y. Improved deep learning technique to detect freezing of gait in Parkinson's disease based on wearable sensors. Electronics, 2020, no. 9(11), pp. 1-12.Li B., Yao Z., Wang J., Wang S., Yang X., Sun Y. Improved deep learning technique to detect freezing of gait in Parkinson's disease based on wearable sensors. Electronics, 2020, no. 9(11), pp. 1-12.

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