As AI tools become embedded in clinical workflows—from note drafting to radiology reads—early evidence suggests that over-reliance may erode clinical skills, a phenomenon known as de-skilling. A 2025 study in The Lancet Gastroenterology & Hepatology found that endoscopists who routinely used AI-assisted adenoma detection experienced a decline in detection rates (29% to 22%) when AI was removed. The finding suggests sustained AI exposure may weaken independent diagnostic performance. Cognitive psychology research points to cognitive off-loading as a likely mechanism: when clinicians passively accept AI outputs, analytic reasoning declines. This risk appears especially pronounced among less experienced clinicians. Studies in radiology…
Author: Abhay Panchal
A large Swedish randomized trial (SCREESCO), published in Nature Medicine, evaluated over 278,000 60-year-olds assigned to: Unlike most prior trials, SCREESCO included a true usual-care control arm, allowing direct comparison of real-world screening impact. Key Findings Participants will be followed through 2030 to determine mortality impact.
The Wall Street Journal highlights the development of a wearable gas-sensing device—nicknamed the “Fitbit for farts”—designed to continuously monitor intestinal gas patterns. Worn in underwear, the battery-powered sensor tracks flatus events in real time using miniaturized electronics and edge computing, reflecting broader trends in continuous biometric monitoring. Researchers hope the technology could do for gastroenterology what wearables like the Apple Watch did for cardiology: generate longitudinal, real-world physiologic data outside the clinic. By analyzing gas patterns over time, the device could potentially help diagnose and manage conditions such as IBS and other functional GI disorders, shifting care from episodic reporting…
Dr. Supriya Rao and Dr. Naresh Gunaratnam interview Daniel Tal Mor, co-founder and CEO of Lumen, a health technology company focused on metabolic measurement. Lumen’s handheld device analyzes a breath sample to estimate whether the body is primarily using carbohydrates or fats for fuel and pairs those measurements with app-based guidance designed to support nutrition.
Gastroenterology may be on the verge of shifting from episodic, flare-driven care to continuous, AI-enabled chronic disease management. According to GI leaders interviewed by Becker’s, AI-powered remote monitoring—particularly in inflammatory bowel disease (IBD)—could use patient-reported symptoms and biomarker data (e.g., fecal calprotectin) to predict flares, guide treatment escalation, and prioritize care in real time.
Cloud infrastructure is emerging as a key enabler of AI-driven innovation in gastroenterology, particularly as hospitals struggle to scale advanced diagnostics using legacy on-premise systems. Cloud-based platforms can integrate endoscopy devices, EMRs, and AI algorithms into interoperable ecosystems—supporting real-time polyp detection, faster computation, and seamless software updates without requiring hardware upgrades.
A new study from the University of California San Diego shows that AI models can analyze clinical notes to predict which ulcerative colitis patients with low-grade dysplasia are most likely to develop colorectal cancer. Using data from over 55,000 patients in the U.S. Veterans Affairs system, the automated workflow extracted key risk factors—such as lesion size, number of dysplastic sites, resection completeness, and inflammation severity—from narrative colonoscopy and pathology reports. The model accurately stratified patients into five long-term cancer risk categories, correctly identifying nearly half as low risk, with ~99% avoiding cancer within two years. By turning unstructured clinical documentation…
The third week of February brings a wide-ranging set of updates across GI oncology, reflecting continued progress in both biological stratification and clinical implementation. This week’s highlights span the care continuum—from colorectal cancer screening strategies and liquid biopsy integration to mechanistic insights into BRAFV600E resistance, metabolically defined pancreatic cancer subtypes, and perioperative immunotherapy in gastroesophageal malignancies. Emerging evidence in early-onset colorectal cancer, POLE-mutant GI tumors, and appendiceal adenocarcinoma further underscores how increasingly granular molecular characterization is reshaping long-held clinical assumptions. Collectively, these updates point to a broader shift in GI oncology—from static genomic labeling toward functional profiling, dynamic biomarkers, and…
The FORTE trial is a large randomized clinical study evaluating whether patients with 1–2 non-advanced adenomatous polyps benefit from earlier surveillance colonoscopy at 5 years compared to delaying follow-up until 10 years. While current guidelines recommend repeat colonoscopy within a 5–10 year window for these low-risk patients, there is limited randomized evidence to determine the optimal interval—leading to wide variation in practice and frequent overuse of early surveillance. Sponsored by NRG Oncology and enrolling an estimated 9,500 participants aged 45–70, the study will compare colorectal cancer incidence, advanced adenoma detection, cancer mortality, and late-stage disease outcomes between individuals randomized to…
Multi-cancer early detection (MCED) blood tests continue to face clinical validation challenges after Grail’s Galleri test failed to meet its primary endpoint in the large NHS-Galleri trial. The three-year study, which screened over 142,000 asymptomatic participants aged 50–77, did not demonstrate a statistically significant reduction in late-stage (III–IV) cancer diagnoses compared to standard screening alone—triggering a sharp decline in the company’s share value. While secondary analyses showed a favorable trend toward fewer late-stage cancers and a meaningful reduction in stage IV diagnoses in certain cancer subgroups, the trial outcome raises fresh concerns around the real-world utility of MCED tests for…