Abstract

Parkinson’s disease (PD) is a critical neurological ailment that affects millions of individuals worldwide. A correct diagnosis of Parkinson’s disease is required for effective treatment. Deep learning (DL) algorithms based on various diagnostic methodologies have been developed to detect PD and resolve related diagnostic issues. This research study offers a complete assessment of published surveys and DL-based diagnosis methodologies for PD recognition. The techniques of DL-based diagnostic approaches for PD recognition, such as PD dataset pre-processing, extraction and selection of features, and classification, are all included in this survey. The limitations and benefits of the proposed methods have also been examined and explored. In addition, we discussed the various datasets used to evaluate the suggested PD recognition algorithms in order to better understand these datasets. The model evaluation metrics and cross-validation techniques used by different studies in this domain have also been explored in this survey. In light of the evaluated literature, we also examined hot upcoming research issues and related solutions. Finally, we came up with several trends and areas for future study that will aid progress in automatic disease recognition, particularly in detecting Parkinson’s disease and its implementation in E-healthcare systems.

Full Text

Published Version
Open DOI Link

Get access to 250M+ research papers

Discover from 40M+ Open access, 3M+ Pre-prints, 9.5M Topics and 32K+ Journals.

Sign Up Now! It's FREE

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call