Simulated Feature Fusion for Parkinson’s Disease Prediction Using Non-aligned Public Handwriting and Voice Datasets

Rashmi Kariganurun Bheemarao; Shobana Thippeswamy Sharadha; Madiraju Snigdha; Neichelle Alvina Mendonca; Pranamya Prakash Rayakar; Muskan Gupta; Panchangam Sai Manogna

doi:10.5614/itbj.ict.res.appl.2026.20.3.2

Authors

Rashmi Kariganurun Bheemarao B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA,
Shobana Thippeswamy Sharadha P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA,
Madiraju Snigdha B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA, India
Neichelle Alvina Mendonca B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA
Pranamya Prakash Rayakar B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA
Muskan Gupta B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA
Panchangam Sai Manogna B.M.S. College of Engineering P.O. Box No.: 1908, Bull Temple Road, Bengaluru-560019, KA

DOI:

https://doi.org/10.5614/itbj.ict.res.appl.2026.20.3.2

Keywords:

feature-level fusion, handwriting analysis, machine learning, multimodal fusion, non-aligned datasets, Parkinson's disease, simulated alignment, SMOTE, voice analysis, XGBoost

Abstract

Parkinson?s disease (PD) is a progressive neurodegenerative disorder that affects over 10 million people worldwide, with prevalence increasing due to aging populations. Early symptoms such as tremor, micrographia, bradykinesia, and voice changes often remain undetected until substantial neuronal loss has occurred, leading to irreversible damage and increased treatment costs. This paper presents a comparative study of five machine learning models?Logistic Regression, Random Forest, Support Vector Machine (RBF Kernel), K-Nearest Neighbors, and XGBoost?for Parkinson?s disease prediction using handwriting and voice-based features. Performance was evaluated using accuracy, precision, recall, F1-score, AUC, 5-fold cross-validation, ROC curve analysis, feature importance analysis, and SHAP-based explainability.XGBoost achieved the highest test-set F1-score (0.7407) and AUC (0.8329), while SVM achieved the highest average cross-validation F1-score. Decision-level fusion further improved recall (0.88) while maintaining competitive overall performance. K-Nearest Neighbors produced the lowest overall performance, while XGBoost achieved the best balance between classification accuracy, F1-score, and AUC.

Downloads

Download data is not yet available.

References

Pereira, C.R., Pereira, D.R., Rosa, G.H., Albuquerque, V.H.C., Weber, S.A.T., Hook, C. & Papa, J.P., Handwritten Dynamics Assessment Through Convolutional Neural Networks: An Application to Parkinson?s Disease Identification, Artif. Intell. Med., 87, pp. 67-77, 2018.

Tsanas, A., Little, M.A., McSharry, P. E., Spielman, J. & Ramig, L.O., Novel Speech Signal Processing Algorithms for High-Accuracy Classification of Parkinson?s Disease, IEEE Trans. Biomed. Eng., 59(5), pp. 1264-1271, 2012 .

Zhou, Z.D., Yi, L.X., Wang, D.Q., Lim, T.M. & Tan., E.K., Role l Dopamine in the Pathophysiology of Parkinson?s Disease, Transl. Neurodegener., 12(1), 44, 2023. doi: 10.1186/s40035-023-00378-6.

Dorsey, E., Elbaz, A. & Nichols, E., Global, Regional, and National Burden of Parkinson?s Disease, 1990?2016, Lancet Neurol., 17(11), pp. 939-953, 2018.

Bloem, B.R., Okun, M.S. & Klein, C., Parkinson?s Disease, Lancet, 397(10291), pp. 2284-2303, 2021. doi: 10.1016/S0140-6736(21)00218-X.

Schapira, A.H.V., Recent Developments in Biomarkers in Parkinson Disease, Curr. Opin. Neurol, 28(4), pp. 374-379, 2015.

Findley, L.J., The Economic Impact of Parkinson's Disease. Parkinsonism Relat Disord., 13, S8-S12, 2007. doi: 10.1016/j.parkreldis.2007.06.003.

Thomas, M., Lenka, A. & Pal, P.K., Handwriting Analysis in Parkinson's Disease: Current Status and Future Directions, Mov Disord Clin Pract, 4, pp. 806-818, 2017. doi: 10.1002/mdc3.12552

Pham, T.D., Texture Classification and Visualization of Time Series of Gait Dynamics, Phys. Rev. E, 97(5), 053301, 2018.

Bandini, A., Orlandi, S., Escalante, H.J., Giovannelli, F., Cincotta, M., Reyes-Garcia, C.A., Vanni, P., Zaccara, G. & Manfredi, C.. Analysis of Facial Expressions in Parkinson's Disease Through Video-Based Automatic Methods, J Neurosci Methods, 281, pp. 7-20, 2017. doi: 10.1016/j.jneumeth.2017.02.006.

Islam, M.A., Majumder, M.Z.H., Hussein, M.A., Hossain, K.M., Miah, M.S., A Review of Machine Learning and Deep Learning Algorithms for Parkinson?s Disease Detection Using Handwriting and Voice Datasets, Heliyon, 10(3), e25469, 2024. doi: 10.1016/j.heliyon.2024.e25469

Alshammri, R., Alharbi, G., Alharbi, E. & Almubark, I., Machine Learning Approaches to Identify Parkinson's Disease Using Voice Signal Features, Front. Artif. Intell., 6,1084001, 2023. doi: 10.3389/frai.2023.1084001

Meghashree, M.B., Kumar, K.D., Vinutha, N. & Akash, D., The Multimodal Fusion of Voice, Gait, and Handwriting Detection of Parkinson?s Disease Using Machine Learning, Int. Res. J. Multidiscip. Scope, 6(4), pp. 1062-1074, 2025. doi: 10.47857/irjms.2025.v06i04.06552.

Benredjem, S., Mekhaznia, T., Rawad, A., Turaev, S., Bennour, A., Sofiane, B., Aborujilah, A. & Al Sarem, M., Parkinson?s Disease Prediction: An Attention-Based Multimodal Fusion Framework Using Handwriting and Clinical Data, Diagnostics, 15(1), 4, 2025. doi: 10.3390/diagnostics15010004 .

Babui, Y.A., Yamini, Ch., Narendra, G. & Ali, Sk.B., Parkinson?s Disease Prediction using Handwritten and Voice Dataset, in Proceedings of the International Conference on Computer Science and Communication Engineering (ICCSCE 2025), Adv. Comput. Sci. Res., 124, pp. 1047-1061, Atlantis Press, 2025. doi: 10.2991/978-94-6463-858-5_87.

Little, M., Parkinsons, UCI Machine Learning Repository, 2007.

Pereira, C.R., Pereira, D.R., Silva, F.A., Masieiro, J.P., Weber, S.A.T., Hook, C. & Papa, J.P, HandPD NewSpiral Dataset, 2018. https: //wwwp.fc.unesp.br/?papa/pub/datasets/Handpd/

Chen, T. & Guestrin, C., XGBoost: A Scalable Tree Boosting System, in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, San Francisco, CA, USA, pp. 785-794, Aug, 2016.

Chawla, N.V., Bowyer, K W., Hall, L.O. & Kegelmeyer, W.P., SMOTE: Synthetic Minority Over-Sampling Technique, J. Artif. Intell. Res., 16, pp. 321-357, 2002.

Lundberg S.M. & Lee, S.-I., A Unified Approach to Interpreting Model Predictions, Advances in Neural Information Processing Systems, 30, pp. 4765-4774, 2017.

Cui, C., Yang, H., Wang, Y., Zhao, S., Asad, Z., Coburn, L.A. & Wilson, K.T., Deep Multimodal Fusion of Image and Non-Image Data in Disease Diagnosis And Prognosis: A Review, npj Digit. Med., 6(1), 92, 2023.

Baltrusaitis, T., Ahuja, C. & Morency, L.-P., Multimodal Machine Learning: A Survey and Taxonomy, IEEE Trans. Pattern Anal. Mach. Intell., 41(2), pp. 423-443, Feb. 2019.

Zhao, S., Dai, G., Li, J., Zhu, X., Huang, X., Li, Y., Tan, M., Wang, L., Fang, P., Chen, X., Yan, N. & Liu, H., An Interpretable Model Based on Graph Learning for Diagnosis of Parkinson's Disease with Voice-related EEG, npj Digit. Med., 7(1), 3, 2024. doi: 10.1038/s41746-023-00983-9.

Giancardo, L., Schez-Ferro, A., Arroyo-Gallego, T., Butterworth, I., Mendoza, C.S., Montero, P., Matarazzo, M., Obeso, J.A., Gray, M.L. & San JosEstar, R., Computer Keyboard Interaction as an Indicator of Early Parkinson's Disease, Sci. Rep., 6, 34468, 2016.

doi: 10.1038/srep34468.

Altham, C., Zhang, H. & Pereira, E., Machine Learning for the Detection and Diagnosis of Cognitive Impairment in Parkinson's Disease: A Systematic Review, PLOS ONE, 19(5), e0303644, 2024. doi: 10.1371/journal.pone.0303644.