Introduction

Accurate diagnosis and monitoring of inflammatory bowel diseases (IBD) are essential for effective management and reduction of complications. Intestinal ultrasound (IUS) is a non-invasive and cost-effective imaging modality for IBD assessment, yet its widespread adoption is limited by operator dependency and interobserver variability. Artificial intelligence (AI) offers the potential to overcome these limitations by improving diagnostic accuracy and standardizing interpretation.

Aims & Methods

This study aimed to develop an AI-assisted model for the automatic detection of active inflammation in IUS images, using bowel wall thickness (BWT) as a key marker. We further aimed to create an automatic tool for detecting bowel segments within ultrasound images and to extend its application from single images to cine loops. Grayscale ultrasound images and cine-loops of the terminal ileum were collected from Crohn’s disease patients undergoing IUS at a tertiary center. The dataset included both normal and abnormal BWT and was annotated by an experienced operator. Preprocessing involved anonymization and normalization. Automated segmentation algorithms were used to isolate bowel wall structures, and deep learning techniques, including convolutional neural networks (CNNs) and YOLO (You Only Look Once) object detection, were employed to analyze the images.

Results

The study included 2,735 grayscale ultrasound images and cine-loops of the terminal ileum from patients with Crohn's disease. The dataset was divided into three sets: a training set of 1980 images, a validation set of 220 images, and a test set of 515 images. Our model achieved robust performance metrics, including sensitivity, specificity, and accuracy of 99.1%, 94.6%, and 91.5%, respectively, for BWT detection, with an area under the receiver operating characteristic curve (AUC) of 0.85. Our model was validated on 10 independent cine-loops, comprising 7 abnormal and 3 normal cases. It accurately identified bowel wall thickening (BWT) in all seven abnormal cases, correctly classifying them as abnormal. Additionally, the model successfully confirmed the absence of BWT in all three normal cine-loops, categorizing them appropriately as normal.

Conclusion

This study represents a significant advancement in the application of artificial intelligence to intestinal ultrasound interpretation by integrating automated segmentation, automatic analysis capabilities, and cine-loop evaluation. Our model extends beyond traditional static image analysis, paving the way for real-time and dynamic assessments of bowel inflammation. This approach highlights the potential of our AI tools to reduce operator dependency, improve diagnostic accuracy, and enhance clinical workflows, making IUS more accessible and reliable for clinicians with varying levels of expertise.

Disclosure

Prof. Dan Carter has received speaker’s fees, research support, or consultancy fees from Takeda, Janssen, AbbVie, and Taro

Introduction

The global burden of Inflammatory bowel disease (IBD), encompassing ulcerative colitis (UC) and Crohn’s disease (CD), is increasing. Due to its chronic nature, IBD entails long-term healthcare resource use, contributing to substantial economic and environmental costs. Digestive endoscopy, while a cornerstone of disease monitoring, is associated with high energy consumption, greenhouse gas (GHG) emissions, and medical waste. Intestinal ultrasound (IUS) has emerged as a non-invasive, accurate, and potentially more sustainable alternative.

Aims & Methods

The present study aimed to compare the environmental impacts in terms of CO₂ emissions, energy consumption (kWh), and disposable waste of IUS and endoscopy for therapeutic monitoring in IBD. In this single-center, retrospective observational study conducted at IRCCS Humanitas Research Hospital (Rozzano, Milan), all consecutive IUS and lower gastrointestinal endoscopies performed for therapeutic monitoring in IBD patients between January 2022 and December 2025 were included. Procedures for oncologic surveillance in long-standing disease, isolated rectal involvement (<15 cm from the anal verge), or without confirmed IBD were excluded. Clinical and procedural data were collected. Environmental metrics such as CO₂ emissions, energy consumption (kWh), and disposable waste were estimated using validated conversion factors and compared using appropriate statistical tests (Independent-samples t-test or Mann–Whitney U test).

Results

Endoscopic procedures performed during study period (n=2698) ↓ Exclusion criteria - Endoscopic procedures in trial setting (n=1157) -Endoscopic surveillance in UC (n= 216) -Pouchoscopies & ulcerative proctitis monitoring (n=162) - Operative/ emergency procedures (n=71) ↓ Endoscopic procedures included in the analysis (n=1092) Colonoscopies (n= 710)↵ ↳ Sigmoidoscopies (n=382)

intestinal ultrasounds performed during study period (n=1541) ↓ Exclusion criteria - Indications other than disease monitoring/ patient BMI> 30 Kg/m2 (n= 231) ↓ Intestinal ultrasounds included in the analysis (n=1310) ​

Flowchart 1. Visual representation of the process of selecting procedures to be included in the final analysis. UC: Ulcerative colitis; BMI: Body Mass Index.

During the study period, a total of 1092 endoscopic procedures, comprising 710 colonoscopies and 382 sigmoidoscopies, along with 1310 IUS were performed in 1028 IBD patients (58% UC, 42% CD). Estimated resource use for all endoscopic procedures totaled 1,256 kWh of electricity, 72,080 L of CO₂ emissions, and 585.6 kg of disposable waste, based on procedural averages (~1.5 kWh, 80 L of CO₂, 650 g waste per colonoscopy; ~50% of values for sigmoidoscopies). In contrast, IUS was associated with an estimated 524 kWh of electricity and 196.5 kg of disposable waste (average ~0.4 kWh, almost no CO₂ consumption, 150 g waste per exam). Compared to endoscopy, IUS reduced energy use by 82%, CO₂ emissions by 95%, and waste generation by 66% (P < 0.05 for all comparisons).

Conclusion

While gastrointestinal endoscopy remains indispensable in certain clinical scenarios, its environmental impact is considerable. In contrast, IUS emits almost no equivalent of CO₂, consumes significantly less energy per examination and produces negligible wastes. For therapeutic monitoring in IBD, IUS has demonstrated high diagnostic accuracy with a markedly lower ecological footprint. These findings promote IUS as a sustainable and environmentally friendly alternative for routine IBD monitoring in appropriate scenarios.

Disclosure

A Armuzzi has received consulting and/or advisory board fees from AbbVie, Allergan, Amgen, Arena, Biogen, Bristol-Myers Squibb, Celltrion, Eli-Lilly, Ferring, Galapagos, Gilead, Janssen, MSD, Mylan, Pfizer, Protagonist Therapeutics, Roche, Samsung Bioepis, Sandoz, Takeda, lecture and/or speaker bureau fees from AbbVie, Amgen, Arena, Biogen, Bristol-Myers Squibb, Celltrion, Eli-Lilly, Ferring, Galapagos, Gilead, Janssen, MSD, Mitsubishi Tanabe, Novartis, Pfizer, Roche, Sandoz, Takeda, Tigenix, and research grants from MSD, Pfizer, Takeda and Biogen. A Repici received consulting fee from Medtronic, Fuji and Olympus and has received research grant from Alfa Sigma, Norgine, Boston and Erbe. C Bezzio received speaker’s fees and consulting fees from for Takeda, MSD, Ferring, Abbvie, Galapagos and Janssen. A Dal Buono has received speaker’s fees from Pfizer, MSD, Eli-Lilly, Celltrion, Janssen, and consulting fees from Ferring. R Gabbiadini has received speaker’s fees from Pfizer, MSD, Ferring, Eli-Lilly, Celltrion, Janssen, consulting fees from Pfizer and Abbvie. G. Privitera has received speaker’s fees from Janssen and Alphasigma. B. Masoni and G. Franchellucci declare no conflict of interests.