Skip to content

Combating PTSD with Machine Learning

Featured Talk at 9:05AM EDT

The United States is facing an urgent crisis in the treatment of its military veterans. Post-traumatic stress disorder (PTSD) is a serious health issue assailing our veteran population. There are currently 300,000+ disability claims and appeals pending in the Board of Veteran Affairs’ (BVA) of the U.S. Department of Veterans Affairs. PTSD claims makeup at least 22% of that total. Presently, the timeline from claim to appeal exceeds 4 years. This logjam in the VA claims and appeals can be mitigated by automating this process using machine learning techniques.

All PTSD claims are legal arguments. As such, all legal arguments must have structured reasoning. Using this legal structure, documents can be processed by Python NLP machine learning models. These models can be used to streamline the process of adjudicating the claims and appeals.

Slides: https://bit.ly/DeepLearningPTSD
Paper: https://github.com/deborahanneharris/legal-documents/

Presented by

Deborah Diller Harris