Mental Condition Monitoring Based on Multimodality Biometry

Area of Science:

• Biomedical Engineering
• Occupational Health
• Data Science

Background:

• Mental distress in the workplace is a growing concern.
• Current screening methods rely on subjective questionnaires.
• Objective, scalable screening tools are needed for early detection and intervention.

Purpose of the Study:

• To develop and validate a multimodality system for objective mental distress screening in office environments.
• To correlate objective biomarker data with established mental health questionnaire scores.
• To assess the system's potential for improving workplace mental healthcare.

Main Methods:

• A four-month field study with 39 volunteers using a prototype system.
• Data collection included PC usage patterns, activity tracking (sleep/activity), and brain activity/behavior during a working memory task (optical topography).
• Supervised machine learning and multiple linear regression were used to model mental scores (Brief Job Stress Questionnaire, Kessler 6 scale) based on objective biomarkers.

Main Results:

• The multimodality system achieved strong correlations (0.6-0.8) between estimated and questionnaire-based mental scores.
• Estimation error was within 24%, indicating good model performance.
• Selected optimal multiple linear regression models demonstrated significant predictive power.

Conclusions:

• The developed multimodality system effectively estimates mental distress scores using objective biomarkers.
• The system shows significant potential as a quantitative tool for workplace mental healthcare.
• This approach offers a promising alternative or supplement to traditional subjective mental health assessments.