A Comprehensive Review of Computational Methods for Automatic Prediction of Schizophrenia With Insight Into...

Area of Science:

• Psychiatry
• Computational Linguistics
• Artificial Intelligence

Background:

• Psychiatrists traditionally use language and speech behavior for psychiatric diagnosis.
• Descriptive psychopathology and phenomenology inform the language used to describe abnormal mental states.
• Early research linked formal thought disorder and language disturbances in psychosis and schizophrenia, leading to clinical rating scales.

Purpose of the Study:

• To explore novel computational methods for analyzing free speech to improve psychosis prediction and diagnosis.
• To address limitations of traditional linguistic measures, including subjectivity, time consumption, and lack of cultural consideration.

Main Methods:

• Utilizing computational sciences for automated speech analysis.
• Employing natural language processing (NLP) and acoustic analysis to examine semantic incoherence.
• Integrating machine learning approaches for classification and prediction of psychosis.

Main Results:

• Recent advances show promise in using computational methods to analyze speech for psychosis diagnosis.
• Automated speech analysis, particularly examining semantic incoherence, is a focus of current research.
• Combining NLP, acoustic analysis, and machine learning enhances prediction and classification capabilities.

Conclusions:

• Computational analysis of speech represents a significant advancement in psychiatric diagnosis.
• Automated speech analysis offers a more objective and potentially efficient method for detecting psychosis.
• Future research integrating advanced computational techniques may further refine psychosis prediction and diagnosis.