ABSTRACT

Post-Traumatic Stress Disorder (PTSD) is a devastating mental disorder with tremendous individual and societal costs. Clinicians are in urgent need of methods, tools, and data to efficiently track, assess, and respond to mental health needs throughout the treatment process, while patients need feedback about how to improve their therapy. To address these issues, this project aims to develop a computational assessment toolkit with patient and clinician interfaces. The system is fundamentally interdisciplinary and requires combining novel insights from multiple fields -- ubiquitous computing, human-computer interaction, and machine learning. Findings from this project will advance these fields and will be beneficial beyond PTSD. The system will be deployed at Emory Healthcare Veterans Program, a nationally renowned initiative that treats members of the military with PTSD. The project will also provide training for students in the burgeoning field of intelligent health.

Treatment for PTSD is constrained by data collection and extraction. The data that are available can be subjective and narrow, presenting a constant obstacle in the delivery, practice, training of psychotherapy. This project addresses the challenge by developing a PE Collective Sensing System (PECSS), a toolkit that will sit atop a conventional mHealth app for PTSD (i.e., PTSD Coach, originally developed by the Veterans Health Administration and Department of Defense). Specifically, the project will (1) develop novel, user-tailored sensing systems that allow patient data transfer and information extraction during both imaginal and in-vivo exposure exercises, (2) design interfaces for continuous monitoring for both clinicians and patients, and (3) develop, validate and deploy computational models of heterogeneous, PE related sensor data that will support and facilitate the improvement of treatment delivery and effectiveness. PECSS will allow clinicians to use automated predictions to deliver better therapeutic treatment and individualized feedback, and patients to better understand the progress they are making and how to improve their exposure exercises. The interfaces, databases, and computational models will be publicly accessible on the web. A course on User-Centered Design for Intelligent Health Care will also be developed to expose medical students and computer science students to the growing inersection of these fields and best practices in interdisciplinary research and innovation in a range of computing disciplines. Finally, the project should impact mental health research approaches in the DoD and the VA.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

Please report errors in award information by writing to: awardsearch@nsf.gov.