Automatic schizophrenic discrimination on fNIRS by using complex brain network analysis and SVM.

Area of Science:

• Neuroscience
• Medical Imaging
• Psychiatry

Background:

• Schizophrenia diagnosis relies heavily on subjective assessments, leading to potential misdiagnosis.
• Objective physiological data and standardized analysis methods are lacking for schizophrenia.
• Functional Near-Infrared Spectroscopy (fNIRS) offers a non-invasive method to measure hemodynamic responses.

Purpose of the Study:

• To investigate the potential of fNIRS as an objective biomarker for schizophrenia diagnosis.
• To develop and validate a classification method for discriminating schizophrenia patients from healthy controls using fNIRS data.

Main Methods:

• Constructed prefrontal brain networks using 52-channel fNIRS oxy-hemoglobin (oxy-Hb) signals.
• Applied Complex Brain Network Analysis (CBNA) for feature extraction.
• Trained a Support Vector Machine (SVM) classifier on data from 42 schizophrenia patients and 34 healthy controls performing a one-back memory task.

Main Results:

• The SVM classifier achieved an overall accuracy of 85.5%.
• Specific accuracies included 92.8% for schizophrenia samples and 76.5% for healthy controls.
• fNIRS data revealed distinct prefrontal network patterns between the two groups.

Conclusions:

• fNIRS, combined with appropriate classification methods, demonstrates significant potential as an objective diagnostic biomarker for schizophrenia.
• This approach may help overcome the limitations of subjective diagnostic criteria.
• Further research is warranted to validate these findings in larger, diverse populations.