Crohn's disease: research progress in decoding pathogenic multi-network and precision management of artificial...

Area of Science:

• Gastroenterology and Radiology
• Artificial Intelligence in Medicine
• Medical Imaging Biomarkers

Background:

• Crohn's disease (CD) is a heterogeneous inflammatory bowel disease with rising global burden.
• Current diagnostic and monitoring tools like fecal calprotectin, ASCA, CTE/MRE, and endoscopy have limitations in reproducibility and phenotypic detail.
• Unmet needs exist in stratifying inflammation, mesenteric involvement, fibrosis, and predicting treatment response or surgical risk.

Purpose of the Study:

• To review the potential of AI-enabled radiomics in computed tomography enterography (CTE) and magnetic resonance enterography (MRE) for Crohn's disease management.
• To link multi-network pathobiology to imaging phenotypes and radiomic signatures for a biologically anchored framework.
• To discuss challenges and propose a roadmap for clinical translation of radiomics in CD.

Main Methods:

• Adoption of a multi-network pathogenic framework (genetics, barrier, microbiome, immune, fibrosis, mesenteric inflammation) to understand imaging phenotypes.
• Focus on AI-enabled radiomics workflows in CTE/MRE for phenotype quantification, feature extraction, and model development.
• Organization of evidence within a mechanism-to-phenotype framework linking pathobiology to imaging and radiomic signatures.

Main Results:

• AI-enabled radiomics shows potential as an imaging biomarker for diagnosis, differential diagnosis, inflammatory activity assessment, fibrosis stratification, treatment response prediction, and surgical risk management in CD.
• The mechanism-to-phenotype framework provides a biologically grounded basis for radiomics interpretation and model design.
• Key challenges to clinical translation include inter-center variability, image acquisition differences, segmentation uncertainty, feature robustness, and limited external validation.

Conclusions:

• AI-enabled radiomics in CTE/MRE holds significant promise for advancing precision management in Crohn's disease.
• Integrating radiomics with immunologic multi-omics can build a translatable evidence framework for personalized CD care.
• Addressing challenges in standardization, validation, and interpretability is crucial for successful clinical implementation.