This technology delivers an objective, data‑driven PTSD assessment based on heart activity, addressing the absence of standardized physiological criteria and reducing dependence on subjective self‑reports.
PTSD affects millions of individuals globally, with particularly high prevalence among military and veteran populations and a significant presence in the broader public. Despite its impact, PTSD diagnosis and ongoing assessment still rely largely on subjective self-reporting and clinical judgment, increasing the risk of missed or inaccurate diagnoses and suboptimal care. As healthcare delivery expands across clinical, military, and virtual environments, there is growing demand for objective, scalable solutions that support accurate screening and continuous monitoring. Tools that deliver reliable physiological insights can enable earlier identification, support personalized treatment decisions, and ultimately improve long-term outcomes for individuals living with PTSD.
Emory researchers have developed a system utilizing a novel heart rate-based window segmentation strategy that analyzes RR interval data — the time between two consecutive R-waves — on an electrocardiogram (ECG) from electrocardiography or similar heart activity measurements. It identifies quiescent segments of RR intervals to extract heart rate variability (HRV) features, which reflects overall variability in heartbeats and autonomic nervous system function. These features are compared against a machine learning classifier trained on data from individuals with and without PTSD, enabling determination of PTSD status or severity.
Validated using ECG recordings from human subjects.
Publication Reinertsen, E., et al. (2017). Heart rate-based window segmentation improves accuracy of classifying posttraumatic stress disorder using heart rate variability measures. Physiological Measurement, 38(6), 1061–1076. https://doi.org/10.1088/1361-6579/aa6e9c