Using Real-World Data to Better Understand Inflammatory Bowel Disease (IBD)

Ulcerative colitis (UC) and Crohn’s Disease (CD) are the two main types of Inflammatory Bowel Disease (IBD), which affects up to 3 million people in the U.S. alone with no known cure.
UC and CD are lifelong conditions that cause chronic inflammation in the intestines, leading to pain and a variety of symptoms in those affected. Research continues around root causes, diagnosis, and best treatment protocols. The OM1 PremiOM UC and CD datasets look to support these research efforts.
● The PremiOM UC dataset is a continually updating database of 12,800+ patients prospectively followed with deep clinical, laboratory and other data, such as longitudinal outcomes, Mayo scores, disease flares and remissions, and treatment response.
● The PremiOM CD dataset is a continually updating database of 13,000+ patients prospectively followed with deep clinical, laboratory and other data, such as longitudinal outcomes, CDAI scores, disease flares and remissions, and treatment response.

In order to build these premium datasets for IBD, certain symptoms are required. These symptoms are documented across clinical notes. Large scale text extraction is required to obtain important information such as symptom names, symptom assertions, and dates for symptoms.
The Text Intelligence team at OM1 was able to use John Snow Labs spark-nlp healthcare technology, to help extract hard to get symptoms from millions of these clinical documents.
This presentation will discuss what components of the John Snow Labs technology helped accomplish this task as well as lessons learned from using John Snow Labs technologies.

Connect with us: