Machine Learning Approach for Parkinson’s Disease Detection: A Comparative Study of SVM Kernels on DaTSCAN Data
DOI:
https://doi.org/10.22399/ijcesen.3090Keywords:
Parkinson’s disease, DaTSCAN, Support Vector Machine, RBF kernel, Polynomial kernel, Sigmoid KernelAbstract
Parkinson’s disease (PD) detection using machine learning presents significant potential for improving diagnostic accuracy. This study investigates the classification of PD patients and healthy controls (HC) using Striatal Binding Ratio (SBR) values derived from DaTSCAN imaging data. Initial exploratory analysis, including Principal Component Analysis (PCA), revealed a complex, nonlinear data distribution, prompting the use of models adept at handling such patterns. The study primarily evaluates Support Vector Machines (SVM) with different kernel functions—Radial Basis Function (RBF), Polynomial, and Sigmoid—leveraging their ability to model nonlinear relationships.
Comparative analysis demonstrated that the SVM-RBF kernel outperformed other kernels, achieving 98.12% accuracy. The Polynomial kernel followed with 94.63% accuracy (C=10, degree=3), while the Sigmoid kernel lagged at 91.68%. The superior performance of the RBF kernel underscores its effectiveness in capturing the intricate nonlinear patterns in DaTSCAN SBR data. Furthermore, when benchmarked against Random Forest, Logistic Regression, K-Nearest Neighbors (KNN), and Convolutional Neural Networks (CNN), the SVM-RBF model consistently exhibited the highest classification accuracy. This study establishes that an optimized SVM with an RBF kernel provides a robust and highly accurate machine learning approach for distinguishing PD patients from healthy controls using DaTSCAN data.
References
[1] Jie Mei, Christian Desrosiers, and Johannes Frasnelli (2021). Machine Learning for the Diagnosis of Parkinson’s Disease: A Review of Literature Frontiers in Aging Neuroscience. 13. https://doi.org/10.3389/fnagi.2021.633752.
[2] De Deurwaerdere P, Gaetani S, Vaughan RA. (2020). Old neurochemical markers, new functional directions? J Neurochem. Mar;152(6):623-626. doi: 10.1111/jnc.14929. Epub 2020 PMID: 31917872.
[3] Andrea Brovelli, Bruno Nazarian, Martine Meunier, Driss Boussaoud, (2011). Differential roles of caudate nucleus and putamen during instrumental learning, NeuroImage, Volume 57(4), 1580-1590, ISSN 1053-8119, https://doi.org/10.1016/j.neuroimage.2011.05.059.
[4] Hörtnagl H, Pifl C, Hörtnagl E, Reiner A, Sperk G. (2020). Distinct gradients of various neurotransmitter markers in the caudate nucleus and putamen of the human brain. J Neurochem. 152:650–62. doi: 10.1111/jnc.14897
[5] Abedelahi A, Hasanzadeh H, Hadizadeh H, Joghataie MT, (2013). Morphometric and volumetric study of caudate and putamen nuclei in normal individuals by MRI: Effect of normal aging, gender, and hemispheric differences. Pol J Radiol. 78(3):7-14. doi: 10.12659/PJR.889364. PMID: 24115954; PMCID: PMC3789937.
[6] Dauer W, Przedborski S. (2003). Parkinson's disease: mechanisms and models. Neuron. 11;39(6):889-909. doi: 10.1016/s0896-6273(03)00568-3. PMID: 12971891
[7] Nandan, N., Sanjay Pande, M.S., Raveesh, B.N. et al, (2024). Sensitive Two-Dimensional Photonic Crystal Biosensor for The Detection of Parkinson’s Disease. J Opt. https://doi.org/10.1007/s12596-024-02306-x
[8] Akam T, Walton ME. (2021). What is dopamine doing in model-based reinforcement learning? Curr Opin Behav Sci. 38:74-82. doi: 10.1016/j.cobeha.2020.10.010. PMID: 37082448; PMCID: PMC7614453.
[9] Seifert KD, Wiener JI. (2013). The impact of DaTscan on the diagnosis and management of movement disorders: A retrospective study. Am J Neurodegener Dis. 2(1):29-34. Epub PMID: 23515233; PMCID: PMC3601468.
[10] Arash Yaghoobi, Homa Seyedmirzaei, Marzie Jamaat, Moein Ala, (2024). The role of AI and machine learning in the diagnosis of Parkinson’s disease and atypical parkinsonisms. Parkinsonism and Related Disorders, Volume 126, 106986, ISSN 1353-8020, https://doi.org/10.1016/j.parkreldis.2024.106986 .
[11] Bega D, Kuo PH, Chalkidou A, Grzeda MT, Macmillan T, Brand C, Sheikh ZH, Antonini A(2021). Clinical utility of DaTscan in patients with suspected Parkinsonian syndrome: a systematic review and meta-analysis. NPJ Parkinsons Dis. Vol. 7(1):43. doi: 10.1038/s41531-021-00185-8. PMID: 34031400; PMCID: PMC8144619.
[12] Sourabarna Roy, Tannistha Pal, Swapan Debbarma, (2024). A Comparative Analysis of Advanced Machine Learning Algorithms to diagnose Parkinson's Disease, Procedia Computer Science, Volume 235, 122-131, ISSN 1877-0509, https://doi.org/10.1016/j.procs.2024.04.015 .
[13] Chien CY, Hsu SW, Lee TL, Sung PS, Lin CC, (2020). Using Artificial Neural Network to Discriminate Parkinson's Disease from Other Parkinsonisms by Focusing on Putamen of Dopamine Transporter SPECT Images. Biomedicines. 9(1):12. doi: 10.3390/biomedicines9010012. PMID: 33374377; PMCID: PMC7823797.
[14] Justin Quan, Lin Xu, Rene Xu, Tyrael Tong, and Jean Su, (2019). DaTscan SPECT Image Classification for Parkinson's Disease, ArXiv, abs/1909.04142
[15] Kurmi A, Biswas S, Sen S, Sinitca A, Kaplun D, Sarkar R. (2022). An Ensemble of CNN Models for Parkinson's Disease Detection Using DaTscan Images. Diagnostics (Basel). 12(5):1173. doi: 10.3390/diagnostics12051173. PMID: 35626328; PMCID: PMC9139649.
[16] Khachnaoui H, Khlifa N, Mabrouk R. (2022). Machine Learning for Early Parkinson's Disease Identification within SWEDD Group Using Clinical and DaTSCAN SPECT Imaging Features. J Imaging. 8(4):97. doi: 10.3390/jimaging8040097. PMID: 35448224; PMCID: PMC9032319.
[17] Pavan Rajkumar Magesh, Richard Delwin Myloth, Rijo Jackson Tom, (2020). An Explainable Machine Learning Model for Early Detection of Parkinson's Disease using LIME on DaTSCAN Imagery, Computers in Biology and Medicine, Volume 126, 104041, ISSN 0010-4825, https://doi.org/10.1016/j.compbiomed.2020.104041
[18] Martínez-Murcia, F.J., Górriz, J.M., Ramírez, J., Illán, I.A., Puntonet, C.G. (2013). Texture Features Based Detection of Parkinson’s Disease on DaTSCAN Images. In: Ferrández Vicente, J.M., Álvarez Sánchez, J.R., de la Paz López, F., Toledo Moreo, F.J. (eds) Natural and Artificial Computation in Engineering and Medical Applications. IWINAC 2013. Lecture Notes in Computer Science, vol 7931. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-38622-0_28
[19] Kavitha Paranjothi, Fathima Ghouse, Revathi Vaithiyanathan, (2024). Detection of Parkinson’s Disease on DaTSCAN Image Using Multi-kernel Support Vector Machine, International Journal of Intelligent Engineering and Systems, Vol.17(5), DOI: 10.22266/ijies2024.1031.05
[20] Babita Majhi and Bhanu Prasad, (2022). Deep learning architectures for Parkinson's disease detection using multi-modal features. Comput. Biol. Med. 146. https://doi.org/10.1016/j.compbiomed.2022.105610
[21] Makarious, M.B., Leonard, H.L., Vitale, D. et al., (2022). Multi-modality machine learning predicting Parkinson’s disease. npj Parkinsons Dis. 8, 35. https://doi.org/10.1038/s41531-022-00288-w
[22] Wuwang, Junho Lee, Fouzi Harrou and Ying Sun (2020). Early Detection of Parkinson's Disease Using Deep Learning and Machine Learning, IEEE Access Digital Object Identifier 10.1109/ACCESS.2020.3016062.
[23] Madhusudhana G K, Sanjaypande M B, Raveesh B N, (2021). Prediction of Parkinson's disease using the Parametric and Non-Parametric Machine Learning Models, IJIRT. Volume 8(7). ISSN: 2349-6002
[24] Nandan N, Sanjay Pande M B, Raveesh B N, Rakesh, (2024). A Machine Learning Approach to Analyzing DATSCAN SBR Values for the Detection of Parkinson's Disease, J. Electrical Systems 20-11s. 1119-1131, https://doi.org/10.52783/jes.7383
[25] Klokkaris, Anthony, and Anna Migdalska-Richards. (2024). An Overview of Epigenetic Changes in the Parkinson’s Disease Brain International Journal of Molecular Sciences 25(11). 6168. https://doi.org/10.3390/ijms25116168
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
