Canada Spends Logo

Explore our Budget 2026 analysis

Grant: Canada Postdoctoral Research Award (CPRA) program

University of Calgary

$140,000.00

Improved Prediction of Strictures in Crohn's Disease Using Machine Learning

FY 2025-26

Summary
Background: Crohn's disease (CD) is a chronic inflammatory condition with complications that necessitate hospitalization and surgery in most patients. These include strictures, a narrowing of the bowel with accumulated scar tissue. Surgical rates in CD have remained steady in recent decades despite the introduction of ground-breaking biologic drugs that led to substantial improvements in the number of patients who achieve and stay in remission. The challenges of identifying early-stage strictures are believed to account for this apparent paradox as they are typically only detected once the symptoms (e.g., pain, nausea, vomiting) become noticeable or when end-stage narrowing is seen on diagnostic imaging (DI). At this stage scar tissue is irreversible and treatment changes are unlikely to prevent surgery. Treatments tailored towards individual patients hold immense potential. Notably, advanced proteomics methods now quantify thousands of proteins from a single blood sample. When combined with DI and clinical observations these biomarkers can improve the prediction of strictures in CD. However, interpreting complex data has impeded the adoption of these approaches in the clinic. We will address this unmet need by developing a tool that leverages machine learning (ML) to help clinicians identify patients who have a high risk of developing a stricture. Importantly, it will be free, easy to use, and available online. Predictions will be made from data routinely available in the clinic (DI, blood tests) and advanced proteomics biomarkers. Thus, it would be ready to deploy in the clinic at the project's completion. Research Question: Can advanced proteomics and ML methods be used to identify patients at a high risk of developing a stricture? Potential Impact: This project could provide clinicians with a means of predicting patients at risk of developing strictures earlier and facilitate treatment decisions aimed to prevent surgery by halting stricture progression.
Program
Canada Postdoctoral Research Award (CPRA) program
Location
Alberta, Canada
Recipient Type
Academia
Type
Grant
Source
CIHR Funding Decisions Database

As published

Id
536793
Name
Lesack Kyle
Region
Alberta
Country
Canada
Orgtype
Institution Paid; Research Institution
Instname2
Nutrition, Metabolism and Diabetes
Theme2
Clinical
Prcname2
Canada Postdoctoral Research Award - A
Projecttitle
Improved Prediction of Strictures in Crohn's Disease Using Machine Learning
Programtype2
Fellowship Programs
Orgnameinp2
University of Calgary
Cihramount2
$140,000
Cihrequipment2
$0
Programname2
Canada Postdoctoral Research Award (CPRA) program
Keyworddelim
Clinical Decision Support System; Crohn'S Disease; Diagnostic Imaging; Gastroenterology; Machine Learning; Proteomics
Abstract
Background: Crohn's disease (CD) is a chronic inflammatory condition with complications that necessitate hospitalization and surgery in most patients. These include strictures, a narrowing of the bowel with accumulated scar tissue. Surgical rates in CD have remained steady in recent decades despite the introduction of ground-breaking biologic drugs that led to substantial improvements in the number of patients who achieve and stay in remission. The challenges of identifying early-stage strictures are believed to account for this apparent paradox as they are typically only detected once the symptoms (e.g., pain, nausea, vomiting) become noticeable or when end-stage narrowing is seen on diagnostic imaging (DI). At this stage scar tissue is irreversible and treatment changes are unlikely to prevent surgery. Treatments tailored towards individual patients hold immense potential. Notably, advanced proteomics methods now quantify thousands of proteins from a single blood sample. When combined with DI and clinical observations these biomarkers can improve the prediction of strictures in CD. However, interpreting complex data has impeded the adoption of these approaches in the clinic. We will address this unmet need by developing a tool that leverages machine learning (ML) to help clinicians identify patients who have a high risk of developing a stricture. Importantly, it will be free, easy to use, and available online. Predictions will be made from data routinely available in the clinic (DI, blood tests) and advanced proteomics biomarkers. Thus, it would be ready to deploy in the clinic at the project's completion. Research Question: Can advanced proteomics and ML methods be used to identify patients at a high risk of developing a stricture? Potential Impact: This project could provide clinicians with a means of predicting patients at risk of developing strictures earlier and facilitate treatment decisions aimed to prevent surgery by halting stricture progression.
Competitiondate
202509
Approvedterm2
2 yrs 0 mth
Pinamesdelim
Lesack, Kyle
Approvedterm
2 ans 0 mois
Cihrcontribution2
~ $140,000~ $0~
University of Calgary - $140,000 Grant (2025-26) | Canada Spends