SBIR-STTR Award

Epidemiological estimates indicate that at least 20 percent of the forces deployed in or returning from the Iraq and Afghanistan wars suffer PTSD or major depression. The number of veterans receiving mental health care has increased from 900,000 to 1.2 million in the past four years. Last year over 400,000 were diagnosed with PTSD, and studies indicate that not all cases of PTSD, depression, and contemplations of suicide are identified using post-deployment mental health assessments. In order to ensure effective and timely treatment, we propose to develop the Clinical Online PTSD and TBI Analysis for Decision Support (COPTADS) system featuring: • Highly Accurate Detection of Symptoms within a diversity of natural language based medical records • Highly Accurate Classification of Patients as very likely to have PTSD/TBI (or not) • Compelling Explanations of why certain patients are believed to have PTSD or TBI • Mobile Access so that doctors anywhere can get COPTADS assessments on a 24/7 basis • Scalability to ensure real-time processing of millions of medical records The SentiMetrix team has worked extensively on the analysis of natural language documents in order to predict whether subjects of blogs have either PTSD or TBI with 80% accuracy

Keywords:
Ptsd, Tbi, Telehealth, Natural Language Processing, Automated Symptom Detection

It is estimated that 20 percent of returning forces from the Iraq and Afghanistan wars are afflicted with PTSD or major depression. In the last year alone, over 400,000 military veterans have been diagnosed with PTSD. Among troops still in the military, experts estimate that one in five has experienced acute stress, anxiety, depression, or other mental health issues as a result of war-zone deployment. Recent studies indicate that not all cases of PTSD, depression, and contemplations of suicide are identified using mental health assessments given to soldiers following deployment.Building on its extensive pre-existing natural language processing and text analytics portfolio, Sentimetrix has developed the COPTADS (Clinical Online PTSD and TBI Analysis and Decision System) to automatically analyze the content of text documents for signals of PTSD or TBI. In Phase I, multiple scoring algorithms were developed trained and tested, demonstrating accuracy approaching that of human experts.In Phase II, we plan to further improve the COPTADS system, significantly improving its accuracy and scalability.Significant functionality enhancements will allow users (clinicians) to execute temporal queries and monitor trends of individual patients and patient groups, as well as observe their responses to various therapies.