A Methodological and Structural Review of Parkinsons Disease Detection Across Diverse Data Modalities

TL;DR

This methodological survey addresses a critical need for multimodal Parkinson's disease detection by systematically reviewing MRI, gait/pose, sensor, handwriting, speech, EEG, and multimodal fusion studies from 2014 to 2024. It aggregates dataset inventories, preprocessing pipelines, and state-of-the-art ML/DL techniques, highlighting both high-performance results and pervasive limitations such as small sample sizes and lack of standardization. The authors identify gaps in cross-modal integration, real-time applicability, and clinical validation, and they propose future directions including 3D/transformer-based MRI analysis, multi-sensor fusion, transfer learning, and explainable AI to improve robustness and clinical trust. Overall, the work provides actionable guidance for researchers and practitioners aiming to develop scalable, accurate, and clinically relevant PD recognition systems with multimodal data.

Abstract

Parkinsons Disease (PD) is a progressive neurological disorder that primarily affects motor functions and can lead to mild cognitive impairment (MCI) and dementia in its advanced stages. With approximately 10 million people diagnosed globally 1 to 1.8 per 1,000 individuals, according to reports by the Japan Times and the Parkinson Foundation early and accurate diagnosis of PD is crucial for improving patient outcomes. While numerous studies have utilized machine learning (ML) and deep learning (DL) techniques for PD recognition, existing surveys are limited in scope, often focusing on single data modalities and failing to capture the potential of multimodal approaches. To address these gaps, this study presents a comprehensive review of PD recognition systems across diverse data modalities, including Magnetic Resonance Imaging (MRI), gait-based pose analysis, gait sensory data, handwriting analysis, speech test data, Electroencephalography (EEG), and multimodal fusion techniques. Based on over 347 articles from leading scientific databases, this review examines key aspects such as data collection methods, settings, feature representations, and system performance, with a focus on recognition accuracy and robustness. This survey aims to serve as a comprehensive resource for researchers, providing actionable guidance for the development of next generation PD recognition systems. By leveraging diverse data modalities and cutting-edge machine learning paradigms, this work contributes to advancing the state of PD diagnostics and improving patient care through innovative, multimodal approaches.

Figures (11)

• Figure 1: Article selection process procedure.
• Figure 2: Article types: journal, conference, and others.
• Figure 3: Year-wise peer-reviewed publications used in the study.
• Figure 4: Various data modalities based references.
• Figure 5: Example of the PD evaluation across various data modalities.