Differential diagnosis of multiple system atrophy with predominant parkinsonism and Parkinson's disease using neural networks

Authors: Mitsunori Tsuda, Kenta Tsuda, Shingo Asano, Yasushi Kato, Masao Miyazaki

Editor's Choice
Journal of the Neurological Sciences.  REVIEW ARTICLE| VOLUME 470, 123411, March 15, 2025

DOI: https://doi.org/10.1016/j.jns.2025.123411

Highlights

• Neural networks can learn complex data relationships, such as in medical diagnosis.
• Multiple system atrophy-parkinsonian type (MSA-P) has poor diagnostic accuracy.
• We used a neural network to distinguish between MSA-P and Parkinson's disease.
• Our neural network demonstrated good accuracy for clinical diagnosis of MSA-P.

Neural networks (NNs) possess the capability to learn complex data relationships, recognise inherent patterns by emulating human brain functions, and generate predictions based on novel data.  We conducted deep learning utilising an NN to differentiate between Parkinson's disease (PD) and the parkinsonian variant (MSA-P) of multiple system atrophy (MSA).

The distinction between PD and MSA-P in the early stages presents significant challenges. Considering the recently reported heterogeneity and random distribution of lesions in MSA, we performed an analysis employing an NN with voxel-based morphometry data from the entire brain as input variables.

The NN's accuracy in distinguishing MSA-P from PD demonstrates sufficient practicality for clinical application.