Objective: To evaluate the diagnostic performance of a multi-atlas MRI–based machine learning model for differentiating degenerative parkinsonian disorders and to examine how classification accuracy varies according to disease duration.

Background: Despite the different prognosis and response to medication, Parkinson’s disease (PD), multiple system atrophy (MSA) and progressive supranuclear palsy (PSP) can share common clinical manifestations, thus the differential diagnosis can be challenge in clinical practice. Therefore, we applied various machine learning models for the differential diagnosis of degenerative parkinsonian disorders in this study.

Method: We retrospectively collected brain MRI of 913 subjects with parkinsonism (494 with PD, 242 with PSP, and 197 with MSA). In all subjects, the volumes of brain structures based on 12 brain atlases were calculated and used to develop differential diagnostic tool using machine learning. We compared the diagnostic performance among the tools with various machine learning techniques, and the performance of the best model based on the disease duration.

Results: Among the various machine learning techniques, the model with combination of Recursive Feature Elimination with Cross-Validation and Random Forest model (RFECV-RF) demonstrated the highest area under curve (AUC) for the differential diagnosis. The AUC was higher than 0.9 for the prediction of PD vs. others, PSP vs. others, and MSA vs. others. The diagnostic performance was better with longer disease duration from 1 year to 4 years, and the best at 4 years of disease duration.

Conclusion: Our study revealed the changes in the brain structures are different among the degenerative parkinsonian disorders, and these changes can be used for the differential diagnosis. Based on our results, we suggested differential diagnostic tool in parkinsonian disorders using machine learning.

[figure2]

[figure1]

[figure3]

References: 1. Raty V, Kuusimaki T, Majuri J, Vahlberg T, Gardberg M, Noponen T, et al. Stability and Accuracy of a Diagnosis of Parkinson Disease Over 10 Years. Neurology. 2025;104(9):e213499.
2. Quattrone A, Bianco MG, Antonini A, Vaillancourt DE, Seppi K, Ceravolo R, et al. Development and Validation of Automated Magnetic Resonance Parkinsonism Index 2.0 to Distinguish Progressive Supranuclear Palsy-Parkinsonism From Parkinson’s Disease. Mov Disord. 2022;37(6):1272-81.
3. Xiong J, Zhu H, Li X, Hao S, Zhang Y, Wang Z, et al. Auto-Classification of Parkinson’s Disease with Different Motor Subtypes Using Arterial Spin Labelling MRI Based on Machine Learning. Brain Sci. 2023;13(11).
4. Cheng D, Wen J, Liu Y, Ding N, Duan Z, Yang Y, et al. Explainable classification of Parkinson’s disease with different motor subtypes by analyzing the synthetic MRI quantitative parameters of subcortical nuclei. Eur J Radiol. 2025;190:112272.
5. Saeed U, Lang AE, Masellis M. Neuroimaging Advances in Parkinson’s Disease and Atypical Parkinsonian Syndromes. Front Neurol. 2020;11:572976.
6. Gaser C, Dahnke R, Thompson PM, Kurth F, Luders E, The Alzheimer’s Disease Neuroimaging I. CAT: a computational anatomy toolbox for the analysis of structural MRI data. Gigascience. 2024;13.
7. Jost ST, Kaldenbach MA, Antonini A, Martinez-Martin P, Timmermann L, Odin P, et al. Levodopa Dose Equivalency in Parkinson’s Disease: Updated Systematic Review and Proposals. Mov Disord. 2023;38(7):1236-52.
8. Goetz CG, Tilley BC, Shaftman SR, Stebbins GT, Fahn S, Martinez-Martin P, et al. Movement Disorder Society-sponsored revision of the Unified Parkinson’s Disease Rating Scale (MDS-UPDRS): scale presentation and clinimetric testing results. Mov Disord. 2008;23(15):2129-70.
9. Hoehn MM, Yahr MD. Parkinsonism: onset, progression, and mortality. Neurology. 1967;17(5):427-.
10. National Institute of Mental H. NIfTI-1 documentation Internet: National Institute of Mental Health; [Available from: http://nifti.nimh.nih.gov/nifti-1/documentation.
11. Schaefer A, Kong R, Gordon EM, Laumann TO, Zuo XN, Holmes AJ, et al. Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cereb Cortex. 2018;28(9):3095-114.
12. Rolls ET, Huang CC, Lin CP, Feng J, Joliot M. Automated anatomical labelling atlas 3. Neuroimage. 2020;206:116189.
13. Eickhoff SB, Stephan KE, Mohlberg H, Grefkes C, Fink GR, Amunts K, et al. A new SPM toolbox for combining probabilistic cytoarchitectonic maps and functional imaging data. Neuroimage. 2005;25(4):1325-35.
14. Computational Brain Anatomy L. MRI-based digital brain atlases Internet: Computational Brain Anatomy Laboratory; [Available from: https://www.cobralab.ca/atlases.
15. Hammers A, Allom R, Koepp MJ, Free SL, Myers R, Lemieux L, et al. Three-dimensional maximum probability atlas of the human brain, with particular reference to the temporal lobe. Hum Brain Mapp. 2003;19(4):224-47.
16. Neuroimaging Informatics T, Resources C. IBSR: Tool/Resource Info Internet: Neuroimaging Informatics Tools and Resources Clearinghouse; [Available from: https://www.nitrc.org/projects/ibsr.
17. Amunts K, Kedo O, Kindler M, Pieperhoff P, Mohlberg H, Shah NJ, et al. Cytoarchitectonic mapping of the human amygdala, hippocampal region and entorhinal cortex: intersubject variability and probability maps. Anat Embryol (Berl). 2005;210(5-6):343-52.
18. Shattuck DW, Mirza M, Adisetiyo V, Hojatkashani C, Salamon G, Narr KL, et al. Construction of a 3D probabilistic atlas of human cortical structures. Neuroimage. 2008;39(3):1064-80.
19. Mori S, Oishi K, Faria AV. White matter atlases based on diffusion tensor imaging. Curr Opin Neurol. 2009;22(4):362-9.
20. Neuromorphometrics. Neuromorphometrics Internet: Neuromorphometrics; [Available from: https://www.neuromorphometrics.com/.
21. Morel A, Magnin M, Jeanmonod D. Multiarchitectonic and stereotactic atlas of the human thalamus. Journal of Comparative Neurology. 1997;387(4):588-630.
22. Kalavathi P, Prasath VB. Methods on Skull Stripping of MRI Head Scan Images-a Review. J Digit Imaging. 2016;29(3):365-79.
23. Shen Q, Zhao W, Loewenstein DA, Potter E, Greig MT, Raj A, et al. Comparing new templates and atlas-based segmentations in the volumetric analysis of brain magnetic resonance images for diagnosing Alzheimer’s disease. Alzheimers Dement. 2012;8(5):399-406.
24. Slinker B, Glantz S. Multiple regression for physiological data analysis: the problem of multicollinearity. American Journal of Physiology-Regulatory, Integrative and Comparative Physiology. 1985;249(1):R1-R12.
25. Guyon I, Elisseeff A. An Introduction to Variable and Feature Selection Journal of Machine Learning Research, 2003. 2003.
26. Ya Y, Ji L, Jia Y, Zou N, Jiang Z, Yin H, et al. Machine Learning Models for Diagnosis of Parkinson’s Disease Using Multiple Structural Magnetic Resonance Imaging Features. Front Aging Neurosci. 2022;14:808520.
27. Zou H. The adaptive lasso and its oracle properties. Journal of the American statistical association. 2006;101(476):1418-29.
28. Zou H, Hastie T. Regularization and variable selection via the elastic net. Journal of the Royal Statistical Society Series B: Statistical Methodology. 2005;67(2):301-20.
29. Guyon I, Weston J, Barnhill S, Vapnik V. Gene selection for cancer classification using support vector machines. Machine learning. 2002;46(1):389-422.
30. Wade B, Joshi S, Gutman B, Thompson P. Machine Learning on High Dimensional Shape Data from Subcortical Brain Surfaces: A Comparison of Feature Selection and Classification Methods2015. 36-43 p.
31. Qin YY, Hsu JT, Yoshida S, Faria AV, Oishi K, Unschuld PG, et al. Gross feature recognition of Anatomical Images based on Atlas grid (GAIA): Incorporating the local discrepancy between an atlas and a target image to capture the features of anatomic brain MRI. Neuroimage Clin. 2013;3:202-11.
32. Dai P, Xiong T, Zhou X, Ou Y, Li Y, Kui X, et al. The alterations of brain functional connectivity networks in major depressive disorder detected by machine learning through multisite rs-fMRI data. Behav Brain Res. 2022;435:114058.
33. Liu W, Liu M, Yang D, Wang M, Tao T, editors. Automatic diagnosis of autism based on functional magnetic resonance imaging and elastic net. 2020 IEEE 5th Information Technology and Mechatronics Engineering Conference (ITOEC); 2020: IEEE.
34. Michel V, Damon C, Thirion B, editors. Mutual information-based feature selection enhances fMRI brain activity classification. 2008 5th IEEE international symposium on biomedical imaging: From nano to macro; 2008: IEEE.
35. Hall MA. Correlation-based feature selection for machine learning: The University of Waikato; 1999.
36. Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. Advances in neural information processing systems. 2017;30.
37. Zarkali A, Thomas GEC, Zetterberg H, Weil RS. Neuroimaging and fluid biomarkers in Parkinson’s disease in an era of targeted interventions. Nat Commun. 2024;15(1):5661.

« Back to 2026 International Congress

MDS Abstracts - https://www.mdsabstracts.org/abstract/brain-atlas-based-machine-learning-for-parkinsonian-disorders-classification-and-evaluation-of-classification-performance-across-disease-duration/