Assessment of pediatric Crohn's disease activity: validation of the magnetic resonance enterography global score (MEGS) against endoscopic activity score (SES-CD)

Development of a multimodal machine-learning fusion model to non-invasively assess ileal Crohn's disease endoscopic activity

BACKGROUND AND OBJECTIVE

Recurrent attentive non-invasive observation of intestinal inflammation is essential for the proper management of Crohn's disease (CD). The goal of this study was to develop and evaluate a multi-modal machine-learning (ML) model to assess ileal CD endoscopic activity by integrating information from Magnetic Resonance Enterography (MRE) and biochemical biomarkers.

METHODS

We obtained MRE, biochemical and ileocolonoscopy data from the multi-center ImageKids study database. We developed an optimized multimodal fusion ML model to non-invasively assess terminal ileum (TI) endoscopic disease activity in CD from MRE data. We determined the most informative features for model development using a permutation feature importance technique. We assessed model performance in comparison to the clinically recommended linear-regression MRE model in an experimental setup that consisted of stratified 2-fold validation, repeated 50 times, with the ileocolonoscopy-based Simple Endoscopic Score for CD at the TI (TI SES-CD) as a reference. We used the predictions' mean-squared-error (MSE) and the receiver operation characteristics (ROC) area under curve (AUC) for active disease classification (TI SEC-CD≥3) as performance metrics.

RESULTS

121 subjects out of the 240 subjects in the ImageKids study cohort had all required information (Non-active CD: 62 [51%], active CD: 59 [49%]). Length of disease segment and normalized biochemical biomarkers were the most informative features. The optimized fusion model performed better than the clinically recommended model determined by both a better median test MSE distribution (7.73 vs. 8.8, Wilcoxon test, p<1e-5) and a better aggregated AUC over the folds (0.84 vs. 0.8, DeLong's test, p<1e-9).

CONCLUSIONS

Optimized ML models for ileal CD endoscopic activity assessment have the potential to enable accurate and non-invasive attentive observation of intestinal inflammation in CD patients. The presented model is available at https://tcml-bme.github.io/ML_SESCD.html.

Simplified Magnetic Resonance Index of Activity Is Useful for Terminal Ileal but not Colonic Disease in Pediatric Crohn Disease

BACKGROUND

Ileocolonoscopy (IC) detects mucosal inflammation and magnetic resonance enterography (MRE) detects transmural inflammation in Crohn disease (CD). We aimed to evaluate the relationship between the simplified magnetic resonance index of activity (MARIAs) and measures of inflammation by IC in children with newly diagnosed CD.

METHODS

Retrospective review of 140 patients 6-18 years of age with CD who had baseline IC and MRE within 5 weeks of diagnosis. MARIAs was calculated for each intestinal segment (terminal ileum [TI], ascending colon, transverse colon, descending colon, sigmoid colon, rectum), defined as (1 × thickness > 3 mm) + (1 × edema) + (1 × fat stranding) + (2 × ulcers). Sensitivity and specificity were derived using receiver operating characteristic (ROC) curves to compare MARIAs to IC findings.

RESULTS

Using IC as the reference standard, the cutoff MARIAs ≥1 identified TI segments with active inflammation with 84% sensitivity, 73% specificity, 85% positive predictive value (PPV), 70% negative predictive value (NPV), and area under the curve (AUC) 0.782 (95% confidence interval [CI] 0.689-0.876). The cutoff MARIAs ≥2 identified TI segments with severe lesions with 87% sensitivity, 76% specificity, 87% PPV, 76% NPV, and AUC 0.814 (95% CI 0.712-0.916). There was poor sensitivity for all colonic segments.

CONCLUSION

The MARIAs is feasible and accurate in reflecting disease activity in the TI, but not in the colon, in children with newly diagnosed CD. Although the MARIAs may be useful for monitoring TI disease activity over time, full assessment continues to require both IC and MRE.

An Overview of Tools to Score Severity in Pediatric Inflammatory Bowel Disease

Background and Aims: The management of IBD entails the use of various treatments (nutrition, medications, and surgery) in order to induce and maintain remission. The assessment of IBD disease activity is based on a combination of symptoms, clinical findings, imaging, and endoscopic procedures. As in any disease, reliable assessment of disease activity or severity is required in order to plan relevant follow-up, decide on appropriate investigations, determine the best treatment option and subsequently assess response to treatment. It is important for proper documentation, follow-up, assessment of response to treatment and communication, especially in patients with IBD, to talk the same language by using validated and widely used scores for disease activity, endoscopic and radiologic activity, and patient reported outcomes both for clinical practice and research. This review aims to highlight key tools available for the assessment of disease activity or severity in individuals (especially children) with IBD. Methods: A literature search was performed using MEDLINE, Pubmed, and the Cochrane Library with the last search date of August 2020. Tools evaluating disease severity across various aspects (clinical, endoscopic, and radiological) were identified and discussed. Those tools validated and specific for children with IBD were included were available. Results: Over time a number of scoring systems have been developed to quantify clinical, endoscopic and imaging assessments in individuals with IBD. While some are exclusively for children or adults, others appear to have relevance to all age groups. In addition, some tools developed in adult populations are utilized in children, but have not expressly been validated in this age group. Conclusions: Although some available scoring tools are appropriate for children with IBD, others require consideration. The development and use of pediatric-specific tools is relevant and appropriate to optimal care of children and adolescents with IBD.