The development of convolutional neural networks has achieved impressive advances of machine learning in recent years, leading to an increasing use of artificial intelligence (AI) in the field of gastrointestinal (GI) diseases. AI networks have been trained to differentiate benign from malignant lesions, analyze endoscopic and radiological GI images, and assess histological diagnoses, obtaining excellent results and high overall diagnostic accuracy. Nevertheless, there data are lacking on side effects of AI in the gastroenterology field, and high-quality studies comparing the performance of AI networks to health care professionals are still limited. Thus, large, controlled trials in real-time clinical settings are warranted to assess the role of AI in daily clinical practice. This narrative review gives an overview of some of the most relevant potential applications of AI for gastrointestinal diseases, highlighting advantages and main limitations and providing considerations for future development.

The present manuscript aims to review the history, recent advances, evidence, and challenges of artificial intelligence (AI) in colonoscopy. Although it is mainly focused on polyp detection and characterization, it also considers other potential applications (i.e., inflammatory bowel disease) and future perspectives. Some of the most recent algorithms show promising results that are similar to human expert performance. The integration of AI in routine clinical practice will be challenging, with significant issues to overcome (i.e., regulatory, reimbursement). Medico-legal issues will also need to be addressed. With the exception of an AI system that is already available in selected countries (GI Genius; Medtronic, Minneapolis, MN, USA), the majority of the technology is still in its infancy and has not yet been proven to reach a sufficient diagnostic performance to be adopted in the clinical practice. However, larger players will enter the arena of AI in the next few months.

Background: The fecal immunochemical test (FIT) is widely implemented as a first‐line tool in organized colorectal cancer (CRC) screening programs, including Italy. Following a positive FIT, colonoscopy is recommended. Computer‐aided detection (CADe) systems have the potential to enhance adenoma detection, particularly in FIT‐positive populations where identifying advanced adenomas is critical for cancer prevention. This study evaluated the diagnostic performance of CADe‐assisted colonoscopy versus standard colonoscopy (SC) in a FIT‐based screening cohort. Methods: In this multicenter, randomized controlled trial, patients with a positive FIT result were randomized to undergo either CADe‐assisted or standard colonoscopy. The primary endpoint was the advanced adenoma detection rate (AADR). Secondary endpoints included overall adenoma detection rate (ADR), adenomas per colonoscopy (APC), and mean withdrawal time (WT). Results: Of 1077 patients enrolled, 68 were excluded due to inadequate bowel preparation, leaving 1009 patients for analysis (CADe: n = 506; SC: n = 503). AADR was comparable between the groups (21.3% vs. 20.5%, p = 0.794). However, CADe significantly improved ADR (67.6% vs. 59.8%, p = 0.012) and APC (1.82 ± 2.12 vs. 1.34 ± 1.81, p < 0.001). Mean WT was longer in the CADe group (17.10 ± 8.28 min vs. 16.13 ± 8.28 min, p = 0.016). Conclusions: In a FIT‐based organized CRC screening setting, CADe did not enhance detection of AADR with a modest increase in withdrawal time. NCT04441580.

Background Ischemic colitis (IC) is caused by a transient hypo-perfusion of the colon leading to mucosal ulcerations, inflammation, and hemorrhage. The primary aim was to identify predictive factors of endoscopic severity of IC. Secondary endpoints were to show epidemiology, clinical presentation, endoscopic findings, and outcomes of IC. Methods In this single-center retrospective analysis, IC was scored according to endoscopy as: grade 1 (hyperemia, 1 cm superficial, partially confluent ulcers); and grade 3 (deep or diffuse ulcers or necrosis). Then, IC was grouped into low(grade 1) and high-grade (grades 2 and 3). Significant (P=0.1) independent factor of severe IC at univariate analysis were entered into multivariate analysis and considered significant at P<0.05. Results 227 patients (male:female 60:167; mean age 72.7 +/- 16.2 years) were included. IC was scored as grade 1 in 137/227 (60.4%), grade 2 in 62/227 (27.3%), and grade 3 in 28/227 (12.3%) patients. At univariate analysis, age (74.9 vs. 71.3 years; P=0.09), diabetes (14.4% vs. 12.4%; P=0.09), and leukocytosis or creatinine elevation (74.4% vs. 60.6%; P=0.032) were associated with endoscopic high-grade IC. At multivariate analysis, leukocytosis and creatinine levels remained associated with high-grade IC (44.7% vs. 29.9%; odds ratio 1.92, 95% confidence interval 1.073.52; P=0.030). Conclusions Although confounding factors cannot be excluded due to study design and patients' characteristics, leukocytosis and/or creatinine elevation at hospital admission were significantly related with endoscopic high-grade IC and might be used to stratify patients for the need of endoscopy.