Machine learning and deep learning applied to EEG and fNIRS for early autism spectrum disorder diagnosis: a systematic review

Key Points

• Early autism spectrum disorder diagnosis is significantly improved through the application of machine learning techniques.
• Specifically, deep learning algorithms analyzed EEG and fNIRS signals to provide robust diagnostic support.
• This systematic review highlights the efficacy of advanced algorithms in assessing neurophysiological data related to autism.
• The implications call for personalized intervention strategies to be developed based on the neurophysiological evidence gathered.

Abstract