Ensemble Learning-Based Approach for Parkinson’s Disease Detection Using Random Forest and Gradient Boosting on Spiral Drawing Biomarkers

Year 2025, Volume: 8 Issue: 3, 144 - 152, 11.09.2025

Ramiz Görkem Birdal

https://doi.org/10.33187/jmsm.1705745

Abstract

Parkinson’s disease (PD) is a neurodegenerative disorder that most profoundly affects motor capabilities. Machine learning approaches have emerged as potential tools for early diagnosis, accomplished through analysis of impairments exhibited in handwriting and drawing tests. In contrast with previous studies, in this work, ensemble learning approaches are incorporated in PD detection through handwriting tests for assessments of motor capabilities for the first time. Specifically, a composite ensemble approach is utilized, combining Random Forest and Gradient Boosting Classifiers, with a Voting Classifier added for enhancing model robustness and preventing overfitting, allowing for increased generalization to new cases. In addition, a careful analysis of feature importance is conducted, and it is determined that pressure and tilt variation act as key markers for PD-related impairments in motor capabilities. Earlier studies have focused predominantly on analysis of motion path locations (X, Y coordinates); in contrast, in this work, the dynamics of variation in pressure and tilt, factors relatively less focused upon but with increased diagnostic value, are emphasized. In contrast with conventional studies utilizing static handwriting tests, in this work, velocity variation enters into diagnostics, and analysis is conducted of variation in drawing fluidity and rapid motion variation contributing to classification performance.

Keywords

Ensemble learning , Feature engineering , Gradient boosting classifiers , Parkinson’s disease (PD) , Random forest

References

• [1] A. U. Haq, J. P. Li, B. L. Y. Agbley, et al., A survey of deep learning techniques based Parkinson’s disease recognition methods employing clinical data, Expert Syst. Appl., 208 (2022), Article ID 118045. https://doi.org/10.1016/j.eswa.2022.118045
• [2] M. Shaban, Deep learning for Parkinson’s disease diagnosis: a short survey, Computers, 12(3) (2023), Article ID 58. https://doi.org/10.3390/computers12030058
• [3] S. Dixit, K. Bohre, Y. Singh, et al., A comprehensive review on AI-enabled models for Parkinson’s disease diagnosis, Electronics, 12(4) (2023), Article ID 783. https://doi.org/10.3390/electronics12040783
• [4] P. Pradeep, J. Kamalkannan, Comprehensive review of literature on Parkinson’s disease diagnosis, Comput. Biol. Chem., 113 (2024), Article ID 108228. https://doi.org/10.1016/j.compbiolchem.2024.108228
• [5] A. Tsanas, M. Little, P. McSharry, et al., Accurate telemonitoring of Parkinson’s disease progression by non-invasive speech tests, IEEE Trans. Biomed. Eng., 57(4) (2010), 884-93. https://doi.org/10.1109/TBME.2009.2036000
• [6] A. Tsanas, M. A. Little, P. McSharry, et al., Nonlinear speech analysis algorithms mapped to a standard metric achieve clinically useful quantification of average Parkinson’s disease symptom severity, J. R. Soc. Interface, 8(59) (2011), 842–855. https://doi.org/10.1098/rsif.2010.0456
• [7] S. B. Appakaya, R. Sankar, Classification of Parkinson’s disease using pitch synchronous speech analysis, Proceedings of the 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), (2018), 1420–1423. https://doi.org/10.1109/EMBC.2018.8512481
• [8] S. B. Appakaya, S. A. Khoshnevis, E. Sheybani, et al., A novel pitch cycle detection algorithm for telemonitoring applications, Proceedings of the 2020 Wireless Telecommunications Symposium (WTS) (2020), 1-4. https://doi.org/10.1109/WTS48268.2020.9198720
• [9] S. B. Appakaya, R. Pratihar, R. Sankar, Parkinson’s disease classification framework using vocal dynamics in connected speech, Algorithms, 16(11) (2023), Article ID 509. https://doi.org/10.3390/a16110509
• [10] B. E. Sakar, M. E. Isenkul, C. O. Sakar, et al., Collection and analysis of a Parkinson speech dataset with multiple types of sound recordings, IEEE J. Biomed. Health Inform., 17(4) (2013), 828–834, Article ID 2245674. https://doi.org/10.1109/JBHI.2013.2245674
• [11] A. Frid, E. J. Safra, H. Hazan, et al., Computational diagnosis of Parkinson’s disease directly from natural speech using machine learning techniques, Proceedings of the 2014 IEEE International Conference on Software Science, Technology and Engineering (SWSTE), (2014), 50–53. https://doi.org/10.1109/SWSTE.2014.17
• [12] Q. Wang, L. Meng, J. Pang, et al., Characterization of EEG data revealing relationships with cognitive and motor symptoms in Parkinson’s disease: A systematic review, Front. Aging Neurosci., 12 (2020), Article ID 587396. https://doi.org/10.3389/fnagi.2020.587396
• [13] A. Jaramillo-Jimenez, D. A. Tovar-Rios, J. A. Ospina, et al., Spectral features of resting-state EEG in Parkinson’s disease: a multicenter study using functional data analysis, Clin. Neurophysiol., 151 (2023), 28–40. https://doi.org/10.1016/j.clinph.2023.03.363
• [14] S. Lee, R. Hussein, M. J. McKeown, A deep convolutional-recurrent neural network architecture for Parkinson’s disease EEG classification, Proceedings of the 2019 IEEE Global Conference on Signal and Information Processing (GlobalSIP)(2019), 1–4. https://doi.org/10.1109/GlobalSIP45357.2019.8969309
• [15] X. Shi, T. Wang, L. Wang, et al., Hybrid convolutional recurrent neural networks outperform CNN and RNN in task-state EEG detection for Parkinson’s disease, Proceedings of the 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), (2019), 939–944. https://doi.org/10.1109/APSIPAASC47483.2019.9023190
• [16] S. K. Khare, V. Bajaj, U. R. Acharya, PDCNNet: an automatic framework for the detection of Parkinson’s disease using EEG signals, IEEE Sens. J., 21(15) (2021), 17017–17024. https://doi.org/10.1109/JSEN.2021.3080135
• [17] S. Cahoon, F. Khan, M. Polk, et al., Wavelet-based convolutional neural network for Parkinson’s disease detection in resting-state electroencephalography, Proceedings of the 2021 IEEE Signal Processing in Medicine and Biology Symposium (SPMB), (2021), 1–5. https://doi.org/10.1109/SPMB52430.2021.9672279
• [18] M. Filippi, E. Sarasso, F. Agosta, Resting-state functional MRI in Parkinsonian syndromes, Mov. Disord. Clin. Pract., 6(2) (2019), 104–117. https://doi.org/10.1002/mdc3.12730
• [19] Y. Li, P. Liang, X. Jia, et al., Abnormal regional homogeneity in Parkinson’s disease: A resting state fMRI study, Clin. Radiol., 71(1) (2016), e28–e34. https://doi.org/10.1016/j.crad.2015.10.006
• [20] A. F. Wolters, S. C. van de Weijer, A. F. Leentjens, et al., Resting-state fMRI in Parkinson’s disease patients with cognitive impairment: A meta-analysis, Park. Relat. Disord., 62 (2019), 16–27. https://doi.org/10.1016/j.parkreldis.2018.12.016
• [21] A. Kazeminejad, S. Golbabaei, H. Soltanian-Zadeh, Graph theoretical metrics and machine learning for diagnosis of Parkinson’s disease using rs-fMRI, Proceedings of the 2017 Artificial Intelligence and Signal Processing Conference (AISP), (2017), 134–139. https://doi.org/10.1109/AISP.2017.8324124
• [22] C. Luo, W. Song, Q. Chen, et al., Reduced functional connectivity in early-stage drug-naive Parkinson’s disease: A resting-state fMRI study, Neurobiol. Aging, 35(2) (2014), 431–441. https://doi.org/10.1016/j.neurobiolaging.2013.08.018
• [23] X. Zhang, Y. Yang, H.Wang, et al., Deep neural networks with broad views for Parkinson’s disease screening, Proceedings of the 2019 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), (2019), 1018–1022. https://doi.org/10.1109/BIBM47256.2019.8983000
• [24] V.M. Ramirez, V. Kmetzsch, F. Forbes, et al., Deep learning models to study the early stages of Parkinson’s disease, Proceedings of the 2020 IEEE 17th International Symposium on Biomedical Imaging (ISBI), (2020), 1534–1537. https://doi.org/10.1109/ISBI45749.2020.9098529
• [25] T. Bikias, D. Iakovakis, S. Hadjidimitriou, et al., DeepFoG: An IMU-based detection of freezing of gait episodes in Parkinson’s disease patients via deep learning, Front. Robot. AI, 8 (2021), Article ID 537384. https://doi.org/10.3389/frobt.2021.537384
• [26] E. Balaji, D. Brindha, R. Balakrishnan, Supervised machine learning-based gait classification system for early detection and stage classification of Parkinson’s disease, Appl. Soft Comput., 94 (2020), Article ID 106494. https://doi.org/10.1016/j.asoc.2020.106494
• [27] L. Sigcha, N. Costa, I. Pavon, et al., Deep learning approaches for detecting freezing of gait in Parkinson’s disease patients through on-body acceleration sensors, Sensors, 20(7) (2020), Article ID 1895. https://doi.org/10.3390/s20071895
• [28] M. Bachlin, M. Plotnik, D. Roggen, et al., Wearable assistant for Parkinson’s disease patients with the freezing of gait symptom, IEEE Trans. Inf. Technol. Biomed., 14(2) (2009), 436–446. https://doi.org/10.1109/TITB.2009.2036165
• [29] M. Alrawis, S. Al-Ahmadi, F. Mohammad, Bridging modalities: A multimodal machine learning approach for Parkinson’s disease diagnosis using EEG and MRI data, Appl. Sci., 14(9) (2024), Article ID 3883. https://doi.org/10.3390/app14093883
• [30] G.W. Jiji, A. Rajesh, M.M. Lakshmi, Diagnosis of Parkinson’s disease using EEG and fMRI, Multimed. Tools Appl., 82 (2023), 14915–14928. https://doi.org/10.1007/s11042-022-14042-0
• [31] M. E. Isenkul, B. E. Sakar, O. Kursun, Improved spiral test using digitized graphics tablet for monitoring Parkinson’s disease, Proceedings of the 2nd International Conference on e-Health and Telemedicine (ICEHTM), (2014), 171-175.