How the Medical University of South Carolina (MUSC) is using Natural Language Processing to improve clinical care
Social determinants of Health (SDoH) are a top priority of agencies globally such as the World Health Organization (WHO), as well as back here in the U.S. where the Center for Disease Control (CDC) has its own variation of goals per Healthy People 2020. The exact definition of what is included in SDoHs varies - but what remains clear is that they are social factors which impact the health of individuals. These may include a myriad of components, such as: stress, social isolation, employment (or lack of), social support, addiction, food insecurity, transportation issues, etc. SDoHs are primarily found within the clinician narrative in electronic health records (EHR), and are difficult to find when trying to identify individuals to ensure proper care.
Sometimes physicians focus excessively on the ‘medical’ problems and don’t pay enough attention to the context that people live in and the social aspects that influence their health. Our study [utilizing Linguamatics NLP] once again highlights the importance of knowing this information in order to provide patients our very best care.
Researchers at the Medical University of South Carolina (MUSC) wanted to see if Natural Language Processing (NLP), a form of Artificial Intelligence (AI), could assist in this area. The research team identified social isolation from clinical narratives for patients with prostate cancer due to the fact that it was recently determined that treatment related side-effects (e.g. incontinence) often led individuals into social isolation.
Researchers found that by utilizing Linguamatics NLP they were able to demonstrate an impressive 90% precision, 97% recall, and 93% F-measure. Moreover, the review of clinical notes to identify mentions of social isolation, which would have taken multiple months of time for a human to complete, took just seconds with NLP, leading to huge time savings.
When people go to the doctor, they do talk about social isolation and other determinants of health. But you won’t find that in the coded data. You have to look at the clinical notes – that’s where the information is embedded. It would take a human many months to sort through the notes looking for mentions of social isolation. In contrast, the [Linguamatics] NLP software combed through the 55, 516 clinical notes comprising 150,990 documents from 3138 prostate cancer patients in the training data set in just eight seconds.
- Vivienne Zhu, M.D., M.S., MUSC Biomedical Informatics Center (BMIC) 1
Contact us to learn more about how MUSC and other leading academic medical centers, hospitals and healthcare systems are utilizing Linguamatics NLP to mine clinical notes in the EHR and other unstructured health data.
1 Quoted in ScienMag external article