Automated Detection of Parkinson's Disease Based on Multiple Types of Sustained Phonations Using Linear Discriminant Analysis and Genetically Optimized Neural Network

• Published in: IEEE journal of translational engineering in health and medicine Vol. 7; pp. 1 - 10
• Main Authors: Ali, Liaqat, Zhu, Ce, Zhang, Zhonghao, Liu, Yipeng
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.01.2019; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Accuracy; Artificial intelligence; Automation; Classification; Correlation coefficients; deep neural network; Diagnostic systems; Dimensionality reduction; Discriminant analysis; Diseases; Feature extraction; genetic algorithm; Genetic algorithms; Hybrid systems; hyper-parameter optimization; linear discriminant analysis; Machine learning; Neural networks; Optimization; Parkinson's disease; Support vector machines; Testing; Training; Voice; Dimensionality reduction; genetic algorithm; hyper-parameter optimization; linear discriminant analysis; Parkinson’s disease; deep neural network

Objective: Parkinson's disease (PD) is a serious neurodegenerative disorder. It is reported that most of PD patients have voice impairments. But these voice impairments are not perceptible to common listeners. Therefore, different machine learning methods have been developed for automated PD de...