Utilizing computer vision for facial behavior analysis in schizophrenia studies: A systematic review

Area of Science:

• Psychiatry
• Computer Science
• Medical Imaging

Background:

• Schizophrenia diagnosis relies on subjective methods like patient reports and clinical observation.
• Objective diagnostic tools are needed to improve accuracy and reduce bias in schizophrenia assessment.

Purpose of the Study:

• To systematically review the application of computer vision for facial behavior analysis in schizophrenia research.
• To evaluate clinical findings, data processing, and machine learning methods used in these studies.

Main Methods:

• A systematic literature search was conducted using PubMed and Google Scholar for studies published up to December 2021.
• Seventeen relevant studies were included, analyzing data collection, facial behavior recognition models, and statistical methods.

Main Results:

• Facial behaviors show potential in estimating schizophrenia diagnosis and psychotic symptoms.
• Studies varied significantly in data collection, hardware, and analytical techniques, with most exhibiting low reporting quality.
• An increasing trend in publications from 2007-2021 indicates growing research interest.

Conclusions:

• Computer vision offers a promising avenue for more objective schizophrenia evaluation.
• Significant variations in methodologies necessitate the development of standardized practices for future research.
• Further research is crucial to advance the field and validate computer vision techniques for schizophrenia.