When: Tuesday 18th August, 2020
Time: 11:00am EDT; 8:00am PDT; 4:00pm BST; 5:00pm CEST.
Duration: 60 minutes.
As healthcare evolves towards value-based care many healthcare organizations are realizing the value of unstructured data as it relates to patient quality and safety. Many important contributing factors on why patients are improving or worsening are often hidden away in text and require NLP to help extract them. Better patient outcomes are reliant on the combination of unstructured data with the structured to give a true 360-degree view of the patient. This is especially important when considering Social Determinants of Health such as social isolation and food insecurity that may be captured in nurse notes and can help identify high risk patients. Another area patient risks can be identified is incidental findings in various clinical tests such as radiology reports that could indicate early signs of cancer.
Another key area of interest in patient safety is improving clinical documentation by identifying missing diagnoses for chronic diseases such as congestive heart failure, aortic stenosis and chronic obstructive pulmonary disorder. By identifying these clinical details in unstructured notes using NLP, patient lives can be improved and millions can be saved with early detection.
Early detection of issues around drugs such as identifying new or known side effects, and earlier identification of drug resistant bacteria is an additional way NLP can help extract and normalize information. These messages are often in HL7 format and require sophisticated approaches to manage the formatting issues with these tables of results from the lab.