Advances in machine learning for the diagnosis of Pakinson's disease
Parkinson's disease(PD)is the second most common neurodegenerative disease after Alzheimer's disease,and the early diagnosis and intervention are crucial for patients.The review focuses on machine learning for intelligent diagnosis of PD.The common machine learning algorithms in PD diagnosis,specifically convolutional neural networks and long short-term memory networks,are introduced,and their applications in medical image analysis and motor behavior analysis are discussed in details.By comparing relevant domestic and international researches,the advantages and disadvantages of using different imaging and kinematic data for PD diagnosis are analyzed.Finally,the review summarizes and presents a prospect for the application of machine learning in PD diagnosis.
万方数据
维普
