Framework for Smart Electronic Health Record- Linked Predictive Models to Optimize Care for Complex Digestive Diseases
Abstract
We utilize de identified database constructed from the world-wide Military Electronic medical record for datasets (constructed for clinical purposes and not research) specific to patients with Crohn's disease (CD) and acute pancreatitis (AP). Using the collected data we developed Bayesian Network (BN) models which successfully predicted outcomes in CD and AP. We shared our model with the University of Pittsburgh Medical Center (UPMC) to confirm our modeling utilizing a different data set. UPMC also successfully predicted similar but not completely matching outcomes with their data set using the Bayesian Network Modeling.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 01, 2015
- Accession Number
- ADA614941
Entities
People
- John D. Betteridge
Organizations
- Henry M. Jackson Foundation for the Advancement of Military Medicine