Machine learning model integrating transient elastography and clinical data for prediction of graft fibrosis after liver transplantation

doi:10.4254/wjh.120258

Advanced Search

BPG is committed to discovery and dissemination of knowledge

Home / Archive / Volume 18, Issue 6

This Article

(1)

(0)

(1)

Peer-Review Report of This Article

CrossCheck and Google Search of This Article

Academic Rules and Norms of This Article

Supplementary Materials of This Article

Citation of this article

Corresponding Author of This Article

Research Domain of This Article

Article-Type of This Article

Open-Access Policy of This Article

Times Cited Counts in Google of This Article

Journal Information of This Article

Publication Name

World Journal of Hepatology

ISSN

1948-5182

Publisher of This Article

Baishideng Publishing Group Inc, 7041 Koll Center Parkway, Suite 160, Pleasanton, CA 94566, USA

- Full Article (XML)

Retrospective Cohort Study

Copyright: ©Author(s) 2026. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution-NonCommercial (CC BY-NC 4.0) license. No commercial re-use. See permissions. Published by Baishideng Publishing Group Inc.

World J Hepatol. Jun 27, 2026; 18(6): 120258
Published online Jun 27, 2026. doi: 10.4254/wjh.120258

Machine learning model integrating transient elastography and clinical data for prediction of graft fibrosis after liver transplantation

Annabel Koivu, Ghazal Azarfar, Maryam Shojaee, Naomi K T Hlaing, Sameera Rizvi, Divya Sharma, Saba Maleki, Mamatha Bhat

Annabel Koivu, Department of Medicine, University Health Network, Toronto M5G 1V7, Ontario, Canada

Ghazal Azarfar, Maryam Shojaee, Sameera Rizvi, Saba Maleki, Toronto General Hospital Research Institute, University Health Network, Toronto M5G 2C4, Ontario, Canada

Naomi K T Hlaing, Mamatha Bhat, Ajmera Transplant Centre, Toronto General Hospital, University Health Network, Toronto M5G 2N2, Ontario, Canada

Divya Sharma, Department of Biostatistics, Dalla Lana School of Public Health, University of Toronto, Toronto M5T 1P8, Ontario, Canada

Co-first authors: Annabel Koivu and Ghazal Azarfar.

Author contributions: Koivu A performed project investigation and drafted original manuscript; Koivu A, Maleki S, and Bhat M contributed to project administration; Azarfar G contributed to methodology, formal analysis, data curation, data interpretation and manuscript revision; Shojaee M, Hlaing NKT, Rizvi S, Maleki S, and Bhat M contributed to manuscript review and editing; Sharma D contributed to validation; Sharma D and Bhat M contributed to validation; Bhat M contributed to study conceptualization; Koivu A and Azarfar G contributed equal to this work and are co-first authors. All authors have read and approve the final manuscript.

AI contribution statement: No AI tools were used in the preparation of this manuscript. The full manuscript text, including the abstract, introduction, materials and methods, results, discussion, and conclusion, was written entirely by the authors without AI assistance. AI tools were not used for language polishing, translation, data analysis, study design, interpretation of results, or image generation. AI assistance (ChatGPT) was used solely for language editing of the separate response-to-reviewers letter and not for any portion of the manuscript itself. The authors carefully reviewed and verified all content and take full responsibility and accountability for the integrity, accuracy, originality, and scientific validity of the manuscript and all submitted materials.

Institutional review board statement: The study was approved by the Ethics Committee of University Health Network (Approval No. 21-6170.7).

Informed consent statement: The requirement for informed consent was waived by the University Health Network Research Ethics Board due to the retrospective nature of the study and use of de-identified data.

Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.

STROBE statement: The authors have read the STROBE Statement-checklist of items, and the manuscript was prepared and revised according to the STROBE Statement-checklist of items.

Data sharing statement: The datasets analyzed during the study are not publicly available due to institutional privacy regulations governing clinical data but may be available from the corresponding author upon reasonable request with appropriate institutional approvals.

Corresponding author: Annabel Koivu, MD, MSc, PhD, FRCPC, Department of Medicine, University Health Network, 500 University Avenue, Suite 602, Toronto M5G 1V7, Ontario, Canada. annabel.koivu2@uhn.ca

Received: February 24, 2026
Revised: March 15, 2026
Accepted: May 29, 2026
Published online: June 27, 2026
Processing time: 120 Days and 15.3 Hours

Abstract

BACKGROUND

Graft fibrosis compromises graft survival in liver transplant (LT) recipients. Early detection is necessary for prevention and treatment. Biopsy is the gold standard for fibrosis assessment; but its invasiveness prevents frequent monitoring.

AIM

To develop a novel non-invasive machine learning (ML) tool combining clinical data with liver stiffness measurements (LSMs) from transient elastography (TE) for personalized graft fibrosis prediction.

METHODS

We performed a retrospective, single-center study including 197 adult LT patients with TE measurements matched to biopsies between 2014-2023. Variables included patient demographics, comorbidities, laboratory values, and TE-derived LSM as a quantitative imaging-derived parameter. Biopsy-confirmed fibrosis stages were dichotomized as significant (≥ F2) vs no or minimal (F0-1). An extreme gradient boosting (XGBoost) ML model was developed, and feature importance assessed using SHapley Additive exPlanations.

RESULTS

The multimodal XGBoost model achieved an area under the receiver operating characteristic curve of 0.90 (95% confidence interval: 0.79-0.99) in predicting significant graft fibrosis. TE-derived LSM was the most influential predictor, followed by graft age, aspartate aminotransferase levels, age, and body mass index. XGBoost outperformed other conventional ML algorithms. Predictions were generated for a test set. Subgroup analysis showed elevated body mass index was associated with increased LSM and greater variability in TE readings, suggesting reduced reliability of non-invasive fibrosis assessment in obese recipients.

CONCLUSION

A multimodal XGBoost model reliably and accurately diagnosed significant graft fibrosis in LT recipients, providing personalized predictions. Individualized SHapley Additive exPlanations analysis revealed LSM measurement as the strongest fibrosis indicator, with graft and recipient age being the significant contributors among other patients.

Keywords: Artificial intelligence; Liver transplantation; Graft fibrosis prediction; Transient elastography; Machine learning; Liver stiffness measurements

Core Tip: Accurate non-invasive assessment of graft fibrosis after liver-transplantation remains challenging. In this study, we developed an extreme gradient boosting-based machine learning model integrating transient elastography-derived liver stiffness measurements with clinical and laboratory variables to predict clinically significant graft fibrosis. The model demonstrated strong diagnostic performance and highlights the potential of multimodal data integration to improved non-invasive fibrosis assessment in transplant recipients. External validation in multicenter cohorts will be required before clinical implementation.