Smartphone gaze-tracking for accessible psychiatric assessment

Area of Science:

• Neuroscience
• Psychiatry
• Biomedical Engineering

Background:

• Eye movements serve as crucial biomarkers for neurological and psychiatric disorders.
• Traditional eye-tracking methods are often cumbersome, expensive, and confined to laboratory settings.
• Deep learning-based gaze tracking on smartphones presents a scalable, accessible alternative for real-world data collection.

Purpose of the Study:

• To evaluate the feasibility of smartphone-based gaze tracking for psychiatric assessment.
• To compare the efficacy of smartphone gaze tracking against research-grade equipment for detecting schizophrenia.
• To assess the utility of smartphone gaze tracking in classifying depressive symptoms.

Main Methods:

• Two studies were conducted: a clinical investigation of schizophrenia patients and a normative study on healthy college students.
• Gaze data was collected using both iPhones and a research-grade EyeLink eye-tracker for schizophrenia detection.
• Smartphone gaze-tracking data from a university cohort was used to classify depressive symptoms via a free-viewing task.

Main Results:

• Smartphone gaze metrics effectively distinguished between individuals with schizophrenia and healthy controls, achieving an AUC of 87.00% and accuracy of 83.33%.
• Performance of the smartphone model was comparable to the EyeLink benchmark (AUC 87.12%, accuracy 86.67%).
• Classification of depressive symptoms using Android smartphones yielded an AUC of 75.54% and accuracy of 75.79%.

Conclusions:

• Smartphone gaze tracking is a viable tool for real-world psychiatric assessment.
• This technology offers an accessible and privacy-preserving method for monitoring psychiatric conditions and treatment.
• The findings support the integration of smartphone-based eye movement analysis into clinical practice.