Artificial intelligence and data integration to achieve improved outcomes and personalised healthcare in IBD
Inflammatory bowel diseases are complex diseases resulting from the combination of multiple etiological factors, and present major challenges in diagnosis, classification and management. Current treatment strategies are applied in an indiscriminate fashion, though anti-inflammatory drugs only benefit a limited number of patients (ceiling of 30-40%) and therefore leaving large margins for improvement. This “one measure does not fit all” situation has spurred the notion that IBD therapy should be tailored to the individual patient.
The current project aims to improve outcomes and personalise healthcare in IBD. First, I will continue my PhD work on TREM-1 as a robust predictive biomarker for anti-TNF therapies. Subsequently I will integrate various data layers – covering different elements in IBD pathogenesis – and explore whether integrated multi-omic biomarkers can improve patient stratification. Beyond successful therapy, patients need routine monitoring to assess residual disease and identify disease-related complications. Ideally, monitoring tools are reproducible and standardised, which is not feasible with MR enterography (MRE) and intestinal ultrasound (IUS), as both require expert interpretation. In order to facilitate the interpretation of imaging and improve outcomes on the long-term, we aim to develop an artificial intelligence-based imaging analysis pipeline (which will also improve central reading processes in future trials using MRE/IUS to assess endpoints).