Application of Machine-Learning Methods for Differentiating Autoimmune Hepatitis from Metabolic Dysfunction-Associated Steatotic Liver Disease: A Study Based on Clinical and Laboratory Evidence

Authors

  • Mohammad Karami Horestani Assistant Professor of Gastroenterology and Hepatology. Department of Internal Medicine, Clinical Research Development Unit, Hajar Hospital, Shahrekord University of Medical Sciences, Shahrekord. Author

Keywords:

Autoimmune Hepatitis, Metabolic Dysfunction-Associated Steatotic Liver Disease, Machine Learning, Clinical Laboratory Data, Differential Diagnosis

Abstract

Autoimmune hepatitis (AIH) and metabolic dysfunction-associated steatotic liver disease (MASLD) are distinct chronic liver disorders that may present with overlapping biochemical, immunological, and metabolic features, making differential diagnosis challenging. This study aimed to establish an evidence-based machine-learning framework for differentiating AIH from MASLD using routinely available clinical and laboratory characteristics. Relevant evidence was synthesized to identify discriminative features, including aminotransferases, bilirubin, immunoglobulin G, autoantibodies, metabolic characteristics, demographic variables, and indicators of hepatic dysfunction and fibrosis. The available evidence indicates that individual clinical or laboratory variables are insufficient for reliable discrimination because autoimmune markers may occur in MASLD and metabolic abnormalities may coexist with AIH. Multivariable machine-learning approaches can integrate these complementary features and identify complex patterns that may not be adequately captured by conventional single-marker assessment. An interpretable framework based on routinely obtainable variables may therefore provide a practical approach for supporting differential classification. Prospective validation in independently characterized patient cohorts remains necessary before clinical implementation.

References

European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of autoimmune hepatitis. J Hepatol. 2025;83(2):453-504.

Muratori L, Lohse AW, Lenzi M. Diagnosis and management of autoimmune hepatitis. BMJ. 2023;380:e070201.

Flikshteyn B, Amer K, Tafesh Z, Pyrsopoulos NT. Diagnosis of autoimmune hepatitis. Clin Liver Dis. 2024;28(1):37-50.

European Association for the Study of the Liver; European Association for the Study of Diabetes; European Association for the Study of Obesity. EASL-EASD-EASO Clinical Practice Guidelines on the management of metabolic dysfunction-associated steatotic liver disease (MASLD). J Hepatol. 2024;81(3):492-542.

Rinella ME, Lazarus JV, Ratziu V, Francque SM, Sanyal AJ, Kanwal F, et al; NAFLD Nomenclature Consensus Group. A multi-society Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol. 2023;79(6):1542-1556.

Younossi ZM, Paik JM, Stepanova M, Ong J, Alqahtani S, Henry L. Clinical profiles and mortality rates are similar for metabolic dysfunction-associated steatotic liver disease and non-alcoholic fatty liver disease. J Hepatol. 2024;80(5):694-701.

McTeer M, Applegate D, Mesenbrink P, Ratziu V, Schattenberg JM, Bugianesi E, et al; LITMUS Consortium investigators. Machine learning approaches to enhance diagnosis and staging of patients with MASLD using routinely available clinical information. PLoS One. 2024;19(2):e0299487.

Fan R, Yu N, Li G, Arshad T, Liu WY, Wong GLH, et al. Machine-learning model comprising five clinical indices and liver stiffness measurement can accurately identify MASLD-related liver fibrosis. Liver Int. 2024;44(3):749-759.

Wei Q, Li W, He S, Wu H, Xie Q, Peng Y, Zhang X. An exploratory machine learning model for predicting advanced liver fibrosis in autoimmune hepatitis patients: A preliminary study. Ann Hepatol. 2025;30(1):101754.

Gerussi A, Saldanha OL, Cazzaniga G, Verda D, Carrero ZI, Engel B, et al. Deep learning helps discriminate between autoimmune hepatitis and primary biliary cholangitis. JHEP Rep. 2025;7(2):101198.

Jensen ASH, Ytting H, Nielsen AB, et al. Diagnostics of autoimmune hepatitis enabled by non-invasive clinical proteomics. Aliment Pharmacol Ther. 2025;62(9):901-919.

Soria A, Díaz A, Iruzubieta P, Martín-Mateos R, Salcedo-Allende MT, Jiménez-Masip A, et al. Autoantibodies are associated with worse outcomes in MASLD. JHEP Rep. 2025;7(10):101470.

Kilani Y, Aldiabat M, Sirilan KYS, Nasir AB, Madi MY, Syn WK. The impact of metabolic dysfunction-associated steatotic liver disease on autoimmune hepatitis outcomes: A nationwide analysis of 2,880 records. Clin Transl Gastroenterol. 2025;16(11):e00912.

Medina-Morales E, Ismail M, Goyal RM, Marenco-Flores A, Saberi B, Fricker Z, et al. Waitlist and transplant outcomes in patients with metabolic dysfunction-associated steatotic liver disease and autoimmune hepatitis. Liver Int. 2024;44(11):3083-3095.

Engel B, Assis DN, Bhat M, Clusmann J, Drenth JPH, Gerussi A, et al; International Autoimmune Hepatitis Group collaborators; European Reference Network for Rare Liver Diseases. Quo vadis autoimmune hepatitis? Summary of the 5th International Autoimmune Hepatitis Group Research Workshop 2024. JHEP Rep. 2025;7(2):101265.

Master SR, Badrick T, Bietenbeck A, Haymond S. Machine learning in laboratory medicine: Recommendations of the IFCC Working Group. Clin Chem. 2023;69(7):690-698.

Downloads

Published

2026-11-03

Issue

Section

Research article

How to Cite

Application of Machine-Learning Methods for Differentiating Autoimmune Hepatitis from Metabolic Dysfunction-Associated Steatotic Liver Disease: A Study Based on Clinical and Laboratory Evidence. (2026). Scientific Journal of Research Studies in Future Basic Sciences and Medical Sciences, 4(2), 58-71. https://journalhi.com/sci/article/view/466

Most read articles by the same author(s)

Similar Articles

31-38 of 38

You may also start an advanced similarity search for this article.