Prospects for integrating artificial intelligence into the practice of gastroenterologists and hepatologists
Open Access
- 24 October 2026
- journal article
- Published by Remedium, Ltd. in Meditsinskiy sovet = Medical Council
- No. 15,p. 182-192
Abstract
Introduction. Artificial intelligence (AI) in clinical practice refers to a set of digital analysis methods capable of performing tasks that typically require a physician's involvement. Unlike conventional medical software, AI algorithms process vast amounts of historical health data to identify patterns and apply them to analyze new data. Sought-after technologies in this field include machine learning, neural network-based deep learning (DL), and large language models (LLMs). Aim. To evaluate the potential and limitations of AI technologies in gastroenterological practice and to explore their contribution to the optimization and enhancement of the quality of clinical decision-making. Materials and methods . A literature search of PubMed/MEDLINE, eLIBRARY.RU, and Google Scholar databases was conducted for the period from 2021 to August 2026. A total of 204 papers were evaluated, and 97 were included in the review. The search terms were as follows: artificial intelligence, machine learning, deep learning, neural networks, large language models, clinical practice, medical diagnostics, multi-criteria decision analysis, explainable artificial intelligence. A supplementary manual search of the reference lists of relevant reviews was conducted. Results. AI systems can analyze diagnostic images, predict disease outcomes, and computerize document processing. Algorithm training replicates clinical experience: the model can acquire differential diagnosis skills by analyzing endoscopic images. LLMs structure electronic health record (EHR) data and support differential diagnosis. MCDA takes into account patients’ quality of life, their individual characteristics, resource constraints, clinical priorities, and prognosis of the individual course of disease. At the same time, AI lacks clinical reasoning and causal reasoning capabilities; performance of AI models is determined by AI training data quality. Conclusion. AI technologies in healthcare serve as supportive tools for physicians that aid, rather than replace, the clinical reasoning of clinicians and other healthcare professionals. Mandatory validation of the scope and efficacy of AI solutions in accordance with established methodological and regulatory requirements will become a reliable basis for their integration into gastroenterologal practice and medical care standards.Keywords
This publication has 84 references indexed in Scilit:
- A Survey of Large Language ModelsFrontiers of Computer Science, 2026
- Beyond the black box: lessons in explainability from AI in mammographyArtificial Intelligence Review, 2026
- Region‐Based Segmentation of Lymph Node Metastases in Whole‐Slide Images of Colorectal Cancer: A Pilot Clinical StudyCancer Medicine, 2026
- Artificial intelligence assistance narrows the experience gap in endoscopic reporting of gastric lesions: a prospective clinical trialSurgical Endoscopy, 2025
- Evaluation efficacy and accuracy of a real-time computer-aided polyp detection system during colonoscopy: a prospective, multicentric, randomized, parallel-controlled study trialSurgical Endoscopy, 2025
- Automated detection of small hepatocellular carcinoma in cirrhotic livers: applying deep learning to Gd-EOB-DTPA-enhanced MRIAbdominal Radiology, 2025
- Diagnosing and grading gastric atrophy and intestinal metaplasia using semi-supervised deep learning on pathological images: development and validation studyGastric Cancer, 2023
- Non-invasive precise staging of liver fibrosis using deep residual network model based on plain CT imagesInternational Journal of Computer Assisted Radiology and Surgery, 2022
- Development of a Framework Based on Reflective MCDA to Support Patient–Clinician Shared Decision-Making: The Case of the Management of Gastroenteropancreatic Neuroendocrine Tumors (GEP-NET) in the United StatesAdvances in Therapy, 2017
- Automatic Liver and Lesion Segmentation in CT Using Cascaded Fully Convolutional Neural Networks and 3D Conditional Random Fields2016