Research on Multi modal Parkinson's Disease Diagnosis System Based on Deep Learning Model

Jiaying Xu; Chao Qiu

doi:10.18282/po4675

Jiaying Xu Department of Neurology，Tiantai People’s Hospital of Zhejiang Province
Chao Qiu ^* Department of Neurology,The First Affiliated Hospital of Zhejiang Chinese Medical University

Article ID: 4675

DOI: https://doi.org/10.18282/po4675

Keywords: Deep Learning, Big Data, Parkinson’s Disease, Multi modal

Abstract

Parkinson’s disease (PD) is a progressive neurodegenerative diseasecaused by reduced dopamine secretion. Deep learning (DL) has receivedwide-spread attention in PD diagnosis research, and the number of papers publishedin this discipline is also increasing. Compared with common machine learningmethods, PD detection using DL presents more promising results. This articlewill review the application of machine learning and deep learning basedonmultimodal imaging in early diagnosis, classification research, and functional pre-diction of Parkinson’s disease, and explain their limitations, as well as provideprospects for future development trends. To achieve our research objectives, weretrieved and analyzed available research papers from databases such as Scopus, Web of Science, Spring, CNKI, and Wanfang.