The challenge
A large healthcare provider wanted to leverage natural language processing (NLP) to mine caregivers’ notes. It wanted to create more complete patient health histories by capturing all these notes in a structured format, identifying and analyzing any details related to social determinants of health (SDH). The organization needed to ascertain which SDH factors, such as economic stability, education, healthcare system and physical environment, significantly impact patients’ health outcomes, and looked to artificial intelligence and machine learning. This healthcare organization turned to Cognizant's healthcare technology expertise.
Our approach
The provider asked Cognizant to develop a text mining engine that would efficiently analyze caregivers’ notes to not only extract specific words and relevant information but also identify context and meaningful insights. We implemented an algorithm that analyzed 900,000 records from approximately 200,000 patients to generate these critical insights.
Our team created seven NLP models, validated by the healthcare provider’s subject matter experts, to produce results on a regular basis, giving the organization a 360-degree view of each patient with more and better SDH details.
Anonymized records of caregiver notes taken at patient visits or other encounters include all notes related to multiple visits for the same medical issue, such as clinical, pre-and post-operative, and discharge notes for a surgery. Our artificial intelligence data analytics solution also identifies people in need of care for a particular disease and points them to specific outreach programs in their local area.