Researchers created an AI tool that predicts recurrence of Barrett’s esophagus following endoscopic eradication therapies with greater than 90% accuracy.
Key Details
- 1AI model predicts recurrence of Barrett’s esophagus and related dysplasia or cancer after EET.
- 2Model achieved over 90% accuracy using data from more than 2,500 patients.
- 3Key risk factors: longer affected tissue length, higher body weight, older age, more treatment sessions, higher-grade dysplasia.
- 4Recurrence occurred in about 30% of patients, with average return two years after therapy.
- 5Validated on internal and external datasets, showing robust performance across groups.
- 6International collaborations underway to further validate AI tool.
Why It Matters
This AI model could enable risk-stratified, personalized surveillance for Barrett’s esophagus after therapy, improving early cancer detection and reducing unnecessary procedures. If widely validated, the tool may optimize healthcare resource use and patient outcomes in endoscopic GI imaging workflows.
Source
EurekAlert
Related News
•EurekAlert
AI-Driven Decision Support Systems Reviewed for Heart Care Impact
A systematic review explores how AI-based clinical decision support systems can enhance cardiovascular disease management while facing adoption and equity challenges.
•EurekAlert
AI Transforms Early Detection and Prediction in Kidney Disease
Artificial intelligence is increasingly enabling earlier detection and improved prediction of kidney disease progression by leveraging complex clinical and imaging data.
•EurekAlert
Stanford Study: Physicians Plus AI Chatbots Improve Clinical Decision-Making
Stanford-led studies show that physicians supported by AI chatbots match or exceed chatbot-alone performance in nuanced clinical decision-making.