Introduction: The use of intestinal ultrasound (IUS) for the diagnosis and follow-up of inflammatory bowel disease is steadily growing. Although access to educational platforms of IUS is feasible, novice ultrasound operators lack experience in performing and interpreting IUS. An artificial intelligence (AI)–based operator supporting system that automatically detects bowel wall inflammation may simplify the use of IUS by less experienced operators. Our aim was to develop and validate an artificial intelligence module that can distinguish bowel wall thickening (a surrogate of bowel inflammation) from normal bowel images of IUS. Methods: We used a self-collected image data set to develop and validate a convolutional neural network module that can distinguish bowel wall thickening >3 mm (a surrogate of bowel inflammation) from normal bowel images of IUS. Results: The data set consisted of 1008 images, distributed uniformly (50% normal images, 50% abnormal images). Execution of the training phase and the classification phase was performed using 805 and 203 images, respectively. The overall accuracy, sensitivity, and specificity for detection of bowel wall thickening were 90.1%, 86.4%, and 94%, respectively. The network exhibited an average area under the ROC curve of 0.9777 for this task. Conclusions: We developed a machine-learning module based on a pretrained convolutional neural network that is highly accurate in the recognition of bowel wall thickening on intestinal ultrasound images in Crohn’s disease. Incorporation of convolutional neural network to IUS may facilitate the use of IUS by inexperienced operators and allow automatized detection of bowel inflammation and standardization of IUS imaging interpretation.

Endoscopy, histology, and cross-sectional imaging serve as fundamental pillars in the detection, monitoring, and prognostication of inflammatory bowel disease (IBD). However, interpretation of these studies often relies on subjective human judgment, which can lead to delays, intra- and interobserver variability, and potential diagnostic discrepancies. With the rising incidence of IBD globally coupled with the exponential digitization of these data, there is a growing demand for innovative approaches to streamline diagnosis and elevate clinical decision-making. In this context, artificial intelligence (AI) technologies emerge as a timely solution to address the evolving challenges in IBD. Early studies using deep learning and radiomics approaches for endoscopy, histology, and imaging in IBD have demonstrated promising results for using AI to detect, diagnose, characterize, phenotype, and prognosticate IBD. Nonetheless, the available literature has inherent limitations and knowledge gaps that need to be addressed before AI can transition into a mainstream clinical tool for IBD. To better understand the potential value of integrating AI in IBD, we review the available literature to summarize our current understanding and identify gaps in knowledge to inform future investigations.

Background: The increasing adoption of intestinal ultrasound (IUS) for monitoring inflammatory bowel diseases (IBD) by IBD providers has uncovered new challenges regarding standardized image interpretation and limitations as a research tool. Artificial intelligence approaches can help address these challenges. We aim to determine the feasibility of radiomic analysis of IUS images and to determine if a radiomics-based classification model can accurately differentiate between normal and abnormal IUS images. We will also compare the radiomic-based model’s performance to a convolutional neural network (CNN)-based classification model to understand which method is more effective for extracting meaningful information from IUS images. Methods: Retrospectively analyzing IUS images obtained during routine outpatient visits, we developed and tested radiomic-based and CNN-based models to distinguish between normal and abnormal images, with abnormal images defined as bowel wall thickness > 3 mm or bowel hyperemia with modified Limberg score ≥ 1 (both are surrogate markers for inflammation). Model performances were measured by area under the receiver operator curve (AUC). Results: For this feasibility study, 125 images (33% abnormal) were analyzed. A radiomic-based model using XG boost yielded the best classifier model with average test AUC 0.98%, 93.8% sensitivity, 93.8% specificity, and 93.7% accuracy. The CNN-based classification model yielded an average testing AUC of 0.75. Conclusions: Radiomic analysis of IUS images is feasible, and a radiomic-based classification model could accurately differentiate abnormal from normal images. Our findings establish methods to facilitate future radiomic-based IUS studies that can help standardize image interpretation and expand IUS research capabilities.

Small bowel involvement in Crohn’s disease (CD) is frequently proximal to the ileocecal valve and inaccessible by conventional ileocolonoscopy (IC). Small bowel capsule endoscopy (SBCE) is among the prime modalities for assessment of small bowel disease in these patients. Intestinal ultrasound (IUS) is an accurate bedside fast and low-cost diagnostic modality utilized in CD for both diagnosis and monitoring. The aim of this study was to examine the accuracy of IUS in patients with suspected CD after a negative IC, and to evaluate the correlation of IUS with SBCE, inflammatory biomarkers and other cross-sectional imaging techniques. Prospective single center study in which patients with suspected CD underwent IUS and SBCE examinations within 3 days. IUS results were blindly compared with SBCE that served as the gold standard. A post hoc comparison was performed of IUS and SBCE results and available cross-sectional imaging results (computed tomography or magnetic resonance enterography) as well as inflammatory biomarkers if measured. The study cohort was followed for 1 year. In case of discordance between the IUS and SBCE results, the diagnosis at 1 year was reported. Fifty patients were included in the study. The diagnostic yield of both IUS and SBCE for the diagnosis of small bowel CD was 38%. The IUS findings significantly correlated to small bowel inflammation detected by SBCE (r = 0.532, p < 0.001), with fair sensitivity and specificity (72% and 84%). Cross-sectional imaging results significantly correlated to IUS as well (r = 0.46, p = 0.018). Follow up was available in 8 of the 10 cases of discordance between IUS and SBCE. In all of these cases, diagnosis of CD was not fully established at the end of the follow up. The diagnostic yield of CE and IUS for detection of CD in patients with negative ileocolonoscopy was similar. IUS can be a useful diagnostic tool in suspected CD when IC is negative.

Cytomegalovirus (CMV) colitis rarely occurs in immunocompetent patients. We report a case of disabling and life threatening diarrhea in an immunocompetent elderly woman due to CMV colitis. The diagnosis of CMV was based on histological examination of tissues biopsied at colonoscopy, positive CMV antigen and high CMV-IgM titer in peripheral blood samples and a good response to systemic gancyclovir treatment. We conclude that CMV should be considered in the differential diagnosis of colitis in elderly immunocompetent patients.