Advanced EEG-based learning approaches to predict schizophrenia: Promises and pitfalls

作者：

Highlights：

• Machine learning shows promise in SZ onset prediction, detection of psychosis risk, and discrimination from other disorders.

• We review EEG-based machine learning methods to discriminate SZ from healthy, at-risk, and subjects with other disorders.

• We synthesize EEG-based deep learning strategies for schizophrenia classification and risk prediction.

• We discuss their potential and limitations and provide future directions in EEG-based model development.

摘要

•Machine learning shows promise in SZ onset prediction, detection of psychosis risk, and discrimination from other disorders.•We review EEG-based machine learning methods to discriminate SZ from healthy, at-risk, and subjects with other disorders.•We synthesize EEG-based deep learning strategies for schizophrenia classification and risk prediction.•We discuss their potential and limitations and provide future directions in EEG-based model development.

论文关键词：Classification,Deep learning,EEG,Machine learning,Prediction,Schizophrenia

论文评审过程：Received 5 August 2020, Revised 11 December 2020, Accepted 16 February 2021, Available online 19 February 2021, Version of Record 7 March 2021.

论文官网地址：https://doi.org/10.1016/j.artmed.2021.102039