Paper Contents
Abstract
Millions of people worldwide suffer from Parkinson's disease (PD), a progressive neurodegenerative illness that manifests both motor and non-motor symptoms. Effective intervention depends on early and precise detection. Recent developments in PD detection using a variety of modalities, including neuroimaging, speech, handwriting, gait, and wearable sensor data, are reviewed in this survey. It highlights how crucial feature extraction, classification, and preprocessing methods are to improving diagnostic performance. The advantages, drawbacks, and clinical applicability of machine learning, deep learning, and transfer learning methodologies are examined in this paper. New developments indicate that multimodal systems hold increasing promise for increased accuracy. The need for explainable AI, high computational costs, and data imbalance are major obstacles. All things considered, the review emphasizes integrative techniques as a viable path toward scalable and trustworthy PD detection.
Copyright
Copyright © 2025 Manisha K M. This is an open access article distributed under the Creative Commons Attribution License.
