Nana Peng, Mary Yue Wang, Sherlot Juan Song, Jimmy Che-To Lai, Vicki Wing-Ki Hui, Vincent Wai-Sun Wong, Grace Lai-Hung Wong, Mamatha Bhat, Terry Cheuk-Fung Yip
Received February 27, 2026 Accepted July 22, 2026 Published online July 23, 2026
Artificial intelligence (AI), particularly foundation and generative models, is reshaping the practice of hepatology through enhanced knowledge synthesis, quantitative and reproducible analysis of multimodal data, and personalized clinical decision support. This narrative review examines the transition from task-specific discrimination AI to large language models (LLMs), multimodal foundation models, and agentic AI. We synthesize evidence from original and validation studies, clinical evaluations, and benchmark studies, as well as expert reviews and regulatory frameworks across metabolic dysfunction-associated steatotic liver disease, chronic hepatitis B, cirrhosis and portal hypertension, hepatocellular carcinoma, and liver transplantation. LLMs can convert free-text notes into structured data, summarize longitudinal electronic health records, support patient education, and retrieve guideline-based information. Retrieval-augmented generation and agentic AI may improve traceability and workflow support, but current evidence is largely retrospective or proof-of-concept. In digital pathology and imaging, discriminative AI has enabled more quantitative and reproducible histologic scoring and biomarker analysis. Pathology and multimodal foundation models offer transferable representations, report generation, and cross-modal reasoning, but hepatology-specific validation remains limited. Key risks include hallucination, automation bias, domain shift across centers and devices, and inequities due to under-representation of patient subgroups. We outline the future directions for safe AI model deployment based on multimodal foundation models, prospective and federated evaluation, lifecycle governance, and continuous monitoring for performance, calibration, and equity. Most generative AI applications in hepatology remain at the proof-of-concept stage, and rigorous prospective validation with human-in-the-loop oversight is required before clinical integration.
Key challenges in cost-effectiveness analyses of emerging MASLD therapies: adherence, adverse events, cardiometabolic benefits, and age-related uncertainty: Correspondence to editorial on “Evaluating treatment response thresholds for cost-effective treatm Eileen L. Yoon, Jeong-Yeon Cho, Mimi Kim, Huiyul Park, Hye-Lin Kim, Dae Won Jun Clinical and Molecular Hepatology.2026; 32(3): e365. CrossRef
Background/Aims Previous studies have identified a substantial degree of agreement between the non-alcoholic fatty liver disease (NAFLD) and metabolic dysfunction-associated steatotic liver disease (MASLD) populations, but the same notion may not apply to normal-weight patients with a lower cardiometabolic risk burden. This study aims to investigate the cardiometabolic risk factor (CMRF) distributions between normal-weight and overweight/obese MASLD, the agreement between historical NAFLD and MASLD, and to compare the risk of liver-related events (LREs) and all-cause mortality in normal-weight versus overweight or obese MASLD.
Methods This study included participants with steatotic liver disease (SLD) from five cohorts in China (Hong Kong), South Korea, and the United States. Participants were recruited from settings including both hospitals and communities. Individuals were classified into normal-weight and overweight/obese groups.
Results This study included 33,793 participants with SLD from five cohorts, of whom 20,893 and 20,701 patients met the diagnosis of NAFLD and MASLD, respectively. Normal-weight patients with NAFLD demonstrated a lower CMRF distribution compared to those with overweight/obese NAFLD. In the community-based cohorts, the proportions with 0 CMRF ranged from 9.0 to 26.7% among normal-weight NAFLD patients, representing the discrepancy between MASLD and NAFLD definitions. Compared with the overweight/obese MASLD, the normalweight MASLD had increased all-cause mortality (normal-weight vs. overweight/obese, 23.44 and 13.80 per 1,000 person-years; P<0.001) but not LREs (2.81 and 2.59 per 1,000 person-years; P=0.54) in the Hong Kong Clinical Data Analysis and Reporting System cohort.
Conclusions Normal-weight individuals with NAFLD demonstrated a lower distribution of CMRFs, resulting in the incomplete agreement between historical NAFLD and MASLD.
Citations
Citations to this article as recorded by
Challenges in defining MASLD in lean individuals: the impact of the Fatty Liver Index on phenotypic characterisation Sherlot Juan Song, Yiwei Liu, Vincent Wai-Sun Wong, Terry Cheuk-Fung Yip Gut.2026; : gutjnl-2026-338216. CrossRef
Beyond BMI: Reassessing the Prevalence of Obesity in Patients With MASLD Under the Lancet Commission Diagnostic Criteria Ru‐Tao Lin, Ren‐Qiang Zeng, Xu‐Ting Shen, Qin‐Mei Sun, Xin Xin, Jia‐Mei Chen, Yi‐Yang Hu, Qin Feng Diabetes, Obesity and Metabolism.2026; 28(8): 6699. CrossRef
Metabolic Dysfunction-Associated Steatotic Liver Disease and Incretin Receptor Agonists: A Metabolic Approach to Halting Liver Disease Progression Ludovico Abenavoli, Anna Giulia Loricchio, Ivo Lopez, Domenico Morano, Abdulrahman Ismaiel, Dan Lucian Dumitrascu, Francesco Luzza Medicina.2026; 62(5): 986. CrossRef
Estimated Body Fat Percentage and Triglyceride‐Glucose Index for Identifying MASLD in Lean Asian Adults: A Cross‐Sectional Analysis Xiang‐Ran Kong, Ya‐Li Chen, Rui Li, Lu‐Xiang Shang, Sha Sha The Kaohsiung Journal of Medical Sciences.2026;[Epub] CrossRef
MASLD and its lean phenotype Kateřina Janstová, Denisa Kyselová, Pavel Trunečka Vnitřní lékařství.2026; 72(4): 232. CrossRef
Xinrui Jin, Sherlot Juan Song, Jimmy Che-To Lai, Grace Lai-Hung Wong, Alice Pik-Shan Kong, Nana Peng, Xiang Xiao, Vincent Wai-Sun Wong, Terry Cheuk-Fung Yip
Clin Mol Hepatol 2026;32(1):353-367. Published online December 1, 2025
Background/Aims Previous studies suggest that hypothyroidism is associated with metabolic dysfunction-associated steatotic liver disease (MASLD) and its histological severity, but clinical outcome data are largely lacking. We aimed to study the impact of hypothyroidism on liver-related events (LREs).
Methods Patients with MASLD were identified from a territory-wide registry in Hong Kong during 2000–2024. Thyroid status was determined using diagnosis codes and thyroid function tests. The primary outcome, LRE, was defined as a composite of hepatic decompensation, hepatocellular carcinoma, liver transplantation, and liver-related death.
Results A total of 20,478 patients with MASLD were included in the final analysis (mean age 56.4±13.2 years; 43.9% male). At baseline, 18,178 (88.8%) patients were euthyroid, 598 (2.9%) were hyperthyroid, and 1,702 (8.3%) were hypothyroid. Compared with euthyroid patients, both hyperthyroidism and overt hypothyroidism were associated with cirrhosis. At a median follow-up of 4.8 years, 179 patients developed LREs, and 26 died from liver disease. Compared with patients with normal serum thyroid-stimulating hormone (TSH) levels of 0.4–4 mIU/L, those with subclinical (4–10 mIU/L; adjusted time-dependent cause-specific hazard ratio [aCSHR], 2.49; 95% CI, 1.51–4.13) and overt hypothyroidism (>10 mIU/L; aCSHR, 4.91; 95% CI, 1.56–15.47) had an increased risk of LREs. Time-dependent, but not baseline, TSH and thyroid status were associated with LRE risk.
Conclusions Subclinical and overt hypothyroidism are associated with an increased risk of LREs in a dose-dependent manner. The association with time-dependent but not baseline thyroid status underscores the importance of thyroid monitoring and suggests that correction of hypothyroidism may mitigate LRE risk.
Citations
Citations to this article as recorded by
Hypertension and long-term adverse clinical outcomes in MASLD: Sensitivity analyses for unmeasured or uncontrolled confounding Guiying Gao, Xiuhong Wang, Ruizhe Huang, Jing Cao Journal of Hepatology.2026; 84(6): e209. CrossRef
Not just fat: muscle also matters in metabolic dysfunction-associated steatotic liver disease (MASLD) Xinyan Zong, Grace Lai-Hung Wong Hepatology International.2026; 20(3): 550. CrossRef
Dynamic Thyroid Function Assessment May Improve Risk Prediction in Metabolic Dysfunction—Associated Steatotic Liver Disease Patients Gabriela Brenta Clinical Thyroidology®.2026; 38(7): 240. CrossRef
Interest of biochemical monitoring of liver function for patients under anti-tuberculosis treatment Alsayed Ahmad Dana, Chikhi Aicha, Sai Touba, Benzitouni Aicha, Aoues Khadidja, Khelfi Abderrezak Drug and Chemical Toxicology.2026; : 1. CrossRef
Immune escape mechanisms revealed by multifunctional nanoprobes targeting PD‐L1 in hepatic cancer progression Yiteng Meng, Liwen Jing, Yeda Chen, Shenggang Zhan, Feng Xiong, Wenbiao Chen VIEW.2026;[Epub] CrossRef
Enzyme‐responsive 612‐AYR nanoparticles for targeted inhibition of B4GALT3‐mediated ITGB1 glycosylation in hepatocellular carcinoma Xiaohui Yu, Xiaoyan Chen, Hui Guo, Xia Jin, Pingping Tan, Longwu Zeng, Yan He, Junjun Li British Journal of Pharmacology.2026; 183(16): 4989. CrossRef
Correspondence to editorial on “Prevalence of clinically significant liver fibrosis in the general population: A systematic review and meta-analysis” Hee Yeon Kim, Miyoung Choi, Dae Won Jun Clinical and Molecular Hepatology.2025; 31(1): e48. CrossRef
In managing metabolic dysfunction-associated steatotic liver disease, which affects over 30% of the general population, effective noninvasive biomarkers for assessing disease severity, monitoring disease progression, predicting the development of liver-related complications, and assessing treatment response are crucial. The advantage of simple fibrosis scores lies in their widespread accessibility through routinely performed blood tests and extensive validation in different clinical settings. They have shown reasonable accuracy in diagnosing advanced fibrosis and good performance in excluding the majority of patients with a low risk of liver-related complications. Among patients with elevated serum fibrosis scores, a more specific fibrosis and imaging biomarker has proved useful to accurately identify patients at risk of liver-related complications. Among specific fibrosis blood biomarkers, enhanced liver fibrosis is the most widely utilized and has been approved in the United States as a prognostic biomarker. For imaging biomarkers, the availability of vibration-controlled transient elastography has been largely improved over the past years, enabling the use of liver stiffness measurement (LSM) for accurate assessment of significant and advanced fibrosis, and cirrhosis. Combining LSM with other routinely available blood tests enhances the ability to diagnose at-risk metabolic dysfunction-associated steatohepatitis and predict liver-related complications, some reaching an accuracy comparable to that of liver biopsy. Magnetic resonance imaging-based modalities provide the most accurate quantification of liver fibrosis, though the current utilization is limited to research settings. Expanding their future use in clinical practice depends on factors such as cost and facility availability.
Citations
Citations to this article as recorded by
Effect of Antidiabetic Drug Classes on the Risk of Liver-Related Events in Individuals With T2D and MASLD Yu Shi, Seung Up Kim, Terry Cheuk-Fung Yip, Emmanuel Tsochatzis, Salvatore Petta, Atsushi Nakajima, Hannes Hagström, Elisabetta Bugianesi, Wah-Kheong Chan, Jérôme Boursier, Boon-Bee George Goh, Arun J. Sanyal, Manuel Romero-Gomez, José Luis Calleja, Victo Clinical Gastroenterology and Hepatology.2026; 24(3): 723. CrossRef
Correspondence to editorial on “Risk stratification by noninvasive tests in patients with metabolic dysfunction-associated steatotic liver disease” Hye Won Lee, Seung Up Kim Clinical and Molecular Hepatology.2026; 32(1): e87. CrossRef
Fe-Mn-IGF-AMD nanoparticles in accurate detection and treatment of liver cirrhosis via regulating relaxation coefficients and attenuating hepatocyte senescence Baihe Wang, Tingting Ji, Zihao Zhang, Yuyao Duan, Zhiqun Xing, Jiazhi Duan, Songbo Zhao, Yang Jia, Qian Wang Colloids and Surfaces B: Biointerfaces.2026; 257: 115182. CrossRef
Letter to the Editor: Are we overestimating LSM-based HCC risk in MASLD? Addressing immortal time bias Heechul Nam, Sung Won Lee Hepatology.2026; 83(6): E171. CrossRef
Stimuli-responsive nanomedicines for hepatic diseases: mechanism, design, recent advances, and clinical translation Leyi Wang, Xue Zhang, Yinggang Li, Min Zhao, Gang Xu, Zhenyu Duan, Qiyong Gong, Kui Luo Journal of Controlled Release.2026; 390: 114522. CrossRef
Genetic Variants in Liver Cirrhosis: Classifications, Mechanisms, and Implications for Clinical Practice Roshni Pushpa Raghavan, Kirti Theresa Alexander, Shine Sadasivan, Chetan Parmar, Manikandan Kathirvel Journal of Personalized Medicine.2026; 16(1): 29. CrossRef
Risk stratification for hepatocellular carcinoma in metabolic dysfunction-associated steatotic liver disease: Editorial on “High Steatosis-Associated Fibrosis Estimator scores predict hepatocellular carcinoma in viral and non-viral hepatitis and metabolic Ho Soo Chun, Minjong Lee Clinical and Molecular Hepatology.2026; 32(1): 368. CrossRef
Prognostic value of non-invasive fibrosis assessment scores in predicting mortality among individuals with metabolic dysfunction-associated steatotic liver disease Lingjie Wu, Shunling Cai, Zhongbin Lin, Ruilie Chen, Yuanfeng Zhang, Xiaobing Gong BMC Public Health.2026;[Epub] CrossRef
Liver Stiffness Measured by Vibration-Controlled Transient Elastography Predicts Hepatic Decompensation in Patients with Hepatocellular Carcinoma Receiving Systemic Treatments Jaejun Lee, Hyun Yang, Si Hyun Bae, Hee Sun Cho, Pil Soo Sung, Jeong Won Jang, Seung Kew Yoon, Keungmo Yang, Heechul Nam, Chang Wook Kim, Hae Lim Lee, Hee Yeon Kim, Sung Won Lee, Ahlim Lee, Do Seon Song, Seok Hwan Kim, Myung Jun Song, Soon Kyu Lee, Jung H Liver Cancer.2026; : 1. CrossRef
The prognostic significance of cholesterol, high-density lipoprotein and glucose (CHG) index in evaluating all-cause mortality risk in metabolic dysfunction-associated steatotic liver disease (MASLD) individuals: evidence from two cohort studies Jicai Wang, Guangjie Tu, Qiang Tao, Ruixi Li, Hang Zhai, Shengjie Hong, Xianjie Shi, Guangquan Zhang Cardiovascular Diabetology.2026;[Epub] CrossRef
Metabolic Dysfunction‐Associated Steatotic Liver Disease and Obesity: Pathogenesis, Diagnostics, Risk Stratification, and Therapeutic Approach Beom Kyung Kim The Kaohsiung Journal of Medical Sciences.2026;[Epub] CrossRef
Secular Trend in Glycaemic Management in Type 2 Diabetes Patients With and Without Cirrhosis Between 2000 and 2023: A Territory‐Wide Cohort Study Mary Yue Wang, Sherlot Juan Song, Nana Peng, Grace Lai‐Hung Wong, Vincent Wai‐Sun Wong, Jimmy Che‐To Lai, Terry Cheuk‐Fung Yip Alimentary Pharmacology & Therapeutics.2026; 64(3): 356. CrossRef
Risk of HCC in Elderly Patients With MASLD: Proposal of an Effective Surveillance Strategy Young Eun Chon, Kyung‐Do Han, Ju Dong Yang, Ju‐Yeong Park, Hyun‐Seok Kim, Hyung Woong Lee, Tae Seop Lim, Beom Kyung Kim Journal of Gastroenterology and Hepatology.2026;[Epub] CrossRef
Challenges and innovations in MASLD and T2DM: Strengthening personalized medicine with SGLT2 inhibitors: Editorial on “Comparative risk of fibrosis progression with sodium-glucose cotransporter-2 vs. dipeptidyl peptidase-4 inhibitors in metabolic dysfunct Yang-Hsiang Lin, Ching-Chih Hu, Chih-Lang Lin Clinical and Molecular Hepatology.2026; 32(3): 1417. CrossRef
Hepatic fibro-inflammation measured by magnetic resonance imaging iron-corrected T1 predicts extrahepatic cancer risk: a UK Biobank study Hye Yeon Chon, Seok-Jae Heo, SungA Bae, Tae Seop Lim, Ja Kyung Kim Clinical and Molecular Hepatology.2026; 32(3): 1349. CrossRef
Prediction of major liver-related events in the population using prognostic models Fredrik Åberg, Ville Männistö Gastroenterology Report.2025;[Epub] CrossRef
Clinical and Economic Burden of Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) in a Spanish Mediterranean Region: A Population-Based Study Javier Díaz Carnicero, Inma Saurí-Ferrer, Josep Redon, Jorge Navarro, Gonzalo Fernández, Carlos Hurtado, Karine Ferreira, Carolina Alvarez-Ortega, Antón Gómez, Carlos J. Martos-Rodríguez, David Martí-Aguado, Desamparados Escudero, Marta Cedenilla Journal of Clinical Medicine.2025; 14(7): 2441. CrossRef
Non-invasive risk-based surveillance of hepatocellular carcinoma in patients with metabolic dysfunction-associated steatotic liver disease Jimmy Che-To Lai, Boyu Yang, Hye Won Lee, Huapeng Lin, Emmanuel A Tsochatzis, Salvatore Petta, Elisabetta Bugianesi, Masato Yoneda, Ming-Hua Zheng, Hannes Hagström, Jerome Boursier, Jose Luis Calleja, George Boon-Bee Goh, Wah-Kheong Chan, Rocìo Gallego-Du Gut.2025; 74(12): 2050. CrossRef
Risk stratification by noninvasive tests in patients with metabolic dysfunction-associated steatotic liver disease Hye Won Lee, Jae Seung Lee, Mi Na Kim, Beom Kyung Kim, Jun Yong Park, Do Young Kim, Sang Hoon Ahn, Seung Up Kim Clinical and Molecular Hepatology.2025; 31(3): 1018. CrossRef
Non-invasive tests of fibrosis in the management of MASLD: revolutionising diagnosis, progression and regression monitoring Gong Feng, Vincent Wai-Sun Wong, Giovanni Targher, Christopher D Byrne, Ming-Hua Zheng Gut.2025; 74(10): 1741. CrossRef
Relative change rate of liver stiffness measurements predicts the risk of liver decompensation in compensated advanced chronic liver disease Yanqiu Li, Zihang Qiao, Jinze Li, Bingbing Zhu, Yu Lu, Ying Feng, Xianbo Wang Clinical and Experimental Medicine.2025;[Epub] CrossRef
Endohepatology in the Management of Liver Diseases Thomas J. Wang, Ajaypal Singh Seminars in Liver Disease.2025; 45(04): 439. CrossRef
Liver stiffness measurement-based risk score for predicting liver decompensation risk: a single-center retrospective Chinese study Yanqiu Li, Zihang Qiao, Jinze Li, Yongqi Li, Ying Feng, Xianbo Wang Clinical and Experimental Medicine.2025;[Epub] CrossRef
Liver Transplantation in the Era of Metabolic Dysfunction–Associated Fatty Liver Disease: Challenges, Ethical Dilemmas, and Future Directions Said A. Al-Busafi, Mohammed Eslam Transplantology.2025; 6(4): 35. CrossRef
Prevalence of MASLD and fibrosis assessed by transient elastography in U.S. adolescents: insights from NHANES 2017-2023 Jialin Wu, Junlong Huang, Shiyu Cao, Yang Lyu, Peiyao Yu, Tiejun Feng, Bonan Chen, Fuda Xie, Ge Zhang, Kangmin Zhuang, Aimin Li, Ka Fai To, Wei Kang Diabetology & Metabolic Syndrome.2025;[Epub] CrossRef
Preface Seung Up Kim Clinical and Molecular Hepatology.2024; 30(Suppl): S3. CrossRef
Meta‐analysis: Efficacy and safety of fibroblast growth factor 21 analogues for the treatment of non‐alcoholic steatohepatitis and non‐alcoholic steatohepatitis‐related fibrosis Rutao Lin, Jianghua Zhou, Qinmei Sun, Xin Xin, Yiyang Hu, Minghua Zheng, Qin Feng Alimentary Pharmacology & Therapeutics.2024; 59(7): 802. CrossRef
Bariatric intervention improves metabolic dysfunction-associated steatohepatitis in patients with obesity: A systematic review and meta-analysis Juchul Hwang, Hyeyoung Hwang, Hyunjae Shin, Bo Hyun Kim, Seong Hee Kang, Jeong-Ju Yoo, Mi Young Choi, Dong eun Lee, Dae Won Jun, Yuri Cho Clinical and Molecular Hepatology.2024; 30(3): 561. CrossRef
First
Prev
