Project 461752
Machine Learning Design of a Post-ERCP Pancreatitis Clinical Decision Tool (PEP-CADET)
Machine Learning Design of a Post-ERCP Pancreatitis Clinical Decision Tool (PEP-CADET)
Project Information
| Study Type: | Unclear |
| Research Theme: | Clinical |
Institution & Funding
| Principal Investigator(s): | Forbes, Nauzer; Heitman, Steven J |
| Co-Investigator(s): | Barkun, Alan N; Brenner, Darren M; Causada Calo, Natalia S; Chen, Yen-I; Lix, Lisa M; Studts, Christina; Tse, Frances |
| Institution: | University of Calgary |
| CIHR Institute: | Nutrition, Metabolism and Diabetes |
| Program: | |
| Peer Review Committee: | Clinical Investigation - C: Digestive, Endocrine and Excretory Systems |
| Competition Year: | 2022 |
| Term: | 4 yrs 0 mth |
Abstract Summary
Endoscopic retrograde cholangio-pancreatography (ERCP) is a commonly performed procedure to treat conditions of the pancreas and bile ducts. About 50 000 ERCPs are performed in Canada every year, comparable to the number of appendectomies or hip replacements. Post-ERCP pancreatitis (PEP) is the most common risk of ERCP, occurring in over 10% of patients. Over $20M of annual Canadian healthcare spending is directly related to the consequences of PEP. Several evidence-based methods exist to reduce the risk of PEP, but many are costly or impractical. A clinical decision tool that predicts PEP risk for a specific patient and gives the doctor recommendations on which PEP prevention method(s) to use could significantly improve patient outcomes and reduce healthcare spending. In this proposal, we describe the design and testing of such a tool.
No special research characteristics identified
This project does not include any of the advanced research characteristics tracked in our database.