Search

Background and Aim: Colonoscopy is commonly used in screening and surveillance for colorectal cancer. Multiple different guidelines provide recommendations on the interval between colonoscopies. This can be challenging for non-specialist healthcare providers to navigate. Large language models like ChatGPT are a potential tool for parsing patient histories and providing advice. However, the standard GPT model is not designed for medical use and can hallucinate. One way to overcome these challenges is to provide contextual information with medical guidelines to help the model respond accurately to queries. Our study compares the standard GPT4 against a contextualized model provided with relevant screening guidelines. We evaluated whether the models could provide correct advice for screening and surveillance intervals for colonoscopy., Methods: Relevant guidelines pertaining to colorectal cancer screening and surveillance were formulated into a knowledge base for GPT. We tested 62 example case scenarios (three times each) on standard GPT4 and on a contextualized model with the knowledge base., Results: The contextualized GPT4 model outperformed the standard GPT4 in all domains. No high-risk features were missed, and only two cases had hallucination of additional high-risk features. A correct interval to colonoscopy was provided in the majority of cases. Guidelines were appropriately cited in almost all cases., Conclusions: A contextualized GPT4 model could identify high-risk features and quote appropriate guidelines without significant hallucination. It gave a correct interval to the next colonoscopy in the majority of cases. This provides proof of concept that ChatGPT with appropriate refinement can serve as an accurate physician assistant., (© 2023 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd.)

Objective: There is debate on the best method of colorectal cancer (CRC) surveillance in inflammatory bowel disease (IBD). We aimed to examine how gastroenterologists around the world practice CRC surveillance and manage dysplastic lesions in IBD., Methods: A 22-question survey was emailed to gastroenterologists from 34 countries. It included questions on resources for, frequency and mode of CRC surveillance, and management of colorectal dysplasia. Fisher's exact test and logistic regression were used to evaluate the differences among respondents in various domains., Results: There were 217 eligible responses, with most gastroenterologists working in public hospitals (76%), and treating >10 patients with IBD weekly (71%). High-definition white light endoscopy (HDWLE) was available in 93.1% of the centers. The preferred mode of surveillance was HDWLE with dye-spray chromoendoscopy and targeted biopsies (41.2%). Fewer than 50% of physicians reported using chromoendoscopy in >50% of cases, citing time as the limiting factor (73.7%). Of these gastroenterologists 63% infrequently (<25% of cases) performed random biopsies during chromoendoscopy. They would attempt endoscopic mucosal resection for polypoid lesions >10 mm (67.2%), including >20 mm lesions with low grade dysplasia (49.8%), and non-polypoid lesions >10 mm without dysplasia (56.9%). For non-polypoid lesions >20 mm with low- and high-grade dysplasia, referral to expert endoscopists was the preferred option., Conclusion: The preferred method of CRC surveillance was HDWLE with chromoendoscopy and targeted biopsies. Random biopsies were infrequently performed. The uptake of chromoendoscopy for surveillance in practice was low. Physicians varied in their approach in removing endoscopically resectable dysplastic lesions., (© 2021 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and John Wiley & Sons Australia, Ltd.)