Identification of Optimal First- and Last-Line Advanced Treatment Strategies in IBD Using Real-World Data Analyzed by the KEM® Explainable Artificial Intelligence Platform: Proof of Concept from the IBD-Drive Project
Chronic Inflammatory Bowel Diseases (IBD) include Crohn’s disease and ulcerative colitis, two conditions characterized by chronic inflammation of the digestive tract, which can damage the intestine and impair patients’ quality of life. To date, treatments for IBD have only a suspensive effect, and therapeutic strategies are based either on reimbursement criteria or on physician expertise, following a “trial-and-error” approach rather than on validated scientific evidence.
The IBD-Drive project aims to optimize patient care by leveraging real-world healthcare data from 1,190 IBD patients, collected by a tertiary care center. This data, gathered through the IBD-adapted Instamed® platform, includes a detailed history of treatments, clinical variables, and disease activity scores such as the Harvey-Bradshaw Index (HBI), the Partial Mayo Score (PMS), and the Simple Clinical Colitis Activity Index (SCCAI).
The KEM® platform (Knowledge Extraction and Management), based on explainable artificial intelligence (xAI), enables a systematic analysis of relationships among all available variables. Its objective is to identify and characterize subgroups of patients with an increased likelihood of response to different therapeutic sequences, thereby paving the way for more personalized and evidence-based medicine.
This proof of concept demonstrates the capability of the KEM® xAI platform to generate actionable hypotheses for guiding therapeutic decisions in the management of IBD.
These initial results are being strengthened and validated as new patient data are integrated and the follow-up of enrolled patients continues. Thanks to its flexibility, our platform continuously incorporates new therapeutic sequences and the latest treatments as soon as they are used in clinical practice.
These data contribute to building an increasingly robust therapeutic decision-support tool.
IBD-Drive will thus enable the development of a real-world data–driven care software, paving the way for a new era of digital precision medicine.