Biometric Data in Post-Traumatic Stress Disorder Detection: A Scoping Review of Digital Health Applications

Area of Science:

• Digital health
• Biometric markers
• Artificial Intelligence (AI) and Machine Learning (ML)

Background:

• Traditional post-traumatic stress disorder (PTSD) assessment relies on self-reports and interviews, potentially delaying diagnosis and limiting accessibility.
• Digital technologies offer potential for earlier, more objective PTSD detection through biometric markers.

Purpose of the Study:

• To comprehensively map biometric modalities utilized for PTSD detection within digital health applications.
• To identify underutilized biometric markers for PTSD detection.
• To characterize the machine learning (ML)/artificial intelligence (AI) approaches applied and assess the inclusion of sex-related analyses.

Main Methods:

• A systematic literature search was conducted across PubMed, IEEE Xplore, and Google Scholar (2015-2025) using a detailed search string.
• Included peer-reviewed human studies employing biometric data with digital tools and/or ML/AI for PTSD detection.
• Data were categorized by modality, digital platform, population, and performance metrics, with findings synthesized narratively.

Main Results:

• Physiological and neuroimaging signals were the most studied biometric modalities; behavioral and genetic markers were underexplored.
• Wearable devices and mobile platforms were common data capture methods, frequently analyzed using ML.
• Performance reporting was inconsistent, sex-stratified analyses were infrequent, and promising modalities like eye-tracking remain underused.

Conclusions:

• Digital biometric approaches show potential for PTSD detection, but progress is hindered by study design heterogeneity and inconsistent reporting.
• There is a need for standardized reporting guidelines in digital health research for PTSD.
• Future research should focus on multimodal models, real-world validation, and sex-inclusive algorithms for equitable PTSD screening.