Gastroenterology and Artificial Intelligence: 3rd Annual Artificial Intelligence Summit (On-Demand) | September 2021-5- Utilizing AI and Machine Learning in IBD Diagnosis and Prognostication

5- Utilizing AI and Machine Learning in IBD Diagnosis and Prognostication

Pdf Summary

Artificial Intelligence (AI) and machine learning have the potential to revolutionize the diagnosis and prognosis of inflammatory bowel disease (IBD). By incorporating multiple data sources such as age, laboratory values, biomarkers, imaging data, and endoscopy findings, machine learning models can predict treatment response and disease status. AI can aid in various facets of endoscopic assessment in IBD, including identifying and classifying Crohn's disease lesions from capsule endoscopy images. Deep learning models have been trained using frames from capsule endoscopy recordings to accurately identify different pathologies.<br /><br />While AI models lack explainability, class activation maps can highlight the regions that led the model to make a classification, validating its ability to detect clinically relevant endoscopic features. Machine learning algorithms have been used to predict remission status in patients on thiopurines using routine laboratory results. The Mucosal Healing Index, which applies AI to mucosal healing markers, demonstrates high accuracy regardless of the location of the disease.<br /><br />The UChicago IBD Biosensor Study aims to gather data from various sources, such as Fitbit sleep data, physical activity data, heart rate data, and clinical data, to assess disease activity and quality of life in IBD patients. Fitbit step measurements have shown predictive power in identifying disease flares.<br /><br />The integration of AI in IBD management can streamline disease monitoring and management by providing objective disease control. The potential of AI in IBD diagnosis, prognosis, treatment selection, and response monitoring holds promise for improving patient outcomes.

Keywords

Artificial Intelligence

machine learning

inflammatory bowel disease

diagnosis

prognosis

treatment response

endoscopic assessment

Crohn's disease

capsule endoscopy

disease monitoring