Feature-driven machine learning to improve early diagnosis of Parkinson's disease
作者：
Highlights：
• We developed a novel hybrid algorithm for aiding early diagnosis of Parkinson's disease.
• We compared the performance of the hybrid algorithm against commercially available software.
• We also compared the performance of the hybrid algorithm against models from published studies.
• Customised algorithmic implementation can improve the overall classification performance.
• Feature reduction via artificial neural networks increased the performance of the hybrid model.
摘要
•We developed a novel hybrid algorithm for aiding early diagnosis of Parkinson's disease.•We compared the performance of the hybrid algorithm against commercially available software.•We also compared the performance of the hybrid algorithm against models from published studies.•Customised algorithmic implementation can improve the overall classification performance.•Feature reduction via artificial neural networks increased the performance of the hybrid model.
论文关键词：Machine learning,Decision support,Parkinson's disease,Diagnosis,Feature selection,Multi-layer perceptron
论文评审过程：Received 28 January 2018, Revised 31 May 2018, Accepted 1 June 2018, Available online 5 June 2018, Version of Record 18 June 2018.
论文官网地址：https://doi.org/10.1016/j.eswa.2018.06.003
