Deep learning for non-invasive detection of steatosis and fibrosis in MASLD: a multicenter study with a new fibroscan-labelled ultrasound dataset
Abdominal Radiology · 24.11.2025 · 51 (6) · 2906-2918
DOI: 10.1007/s00261-025-05309-9 ↗Kurz gefasst
Multizentrische Studie in der Zeitschrift Abdominal Radiology, online erschienen am 24.11.2025. Verfasst von Kartik Bose und 11 weiteren Autor:innen. Beteiligte Einrichtung: Postgraduate Institute of Medical Education and Research, Chandigarh, India. Hauptthemen laut MEDLINE: Deep Learning, Liver Cirrhosis, Fatty Liver, Image Interpretation, Computer-Assisted. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 23 Literaturangaben. Gefördert von: Science Energy Research Board, Department of Science and Technology, Government of India.
Abstract (Originalsprache)
PURPOSE: This study aimed to develop and validate deep learning models for non-invasive assessment of hepatic steatosis and fibrosis using conventional B-mode ultrasound images, with Fibroscan-derived measurement as reference standard. METHODS: We utilized a multi-source approach comprising three distinct ultrasound datasets of patients with MASLD: a private clinical dataset (DS1, n = 111 patients, 1131 images), and two subsets from the public BEHSOF repository-Behbood Clinic subset (DS2, n = 95 patients, 1328 images) for training/validation, and Taleghani Hospital subset (DS3, n = 18 patients, 185 images) for external testing. Additionally, we validated on a temporally independent test set (DS4, n = 23 patients, 155 images). Deep learning models, including EfficientNet-B4 and Vision Transformers, were trained to classify steatosis (S0-S1 vs. S2-S3) and fibrosis (F0-F1 vs. F2-F4), respectively, using 5-fold cross-validation. Performance was assessed and compared with expert radiologists. RESULTS: For steatosis classification, our models achieved an AUROC of 0.83 (95% CI: 0.78-0.88) in cross-validation and 0.70 (95% CI: 0.43-0.93) on the external test set. For fibrosis, AUROC reached 0.86 (95% CI: 0.81-0.91) in cross-validation and 0.88 (95% CI: 0.71-1.00) on the external test set. On the temporally independent test set, performance was better with AUROCs of 0.82 (95% CI: 0.66-0.98) for steatosis and 0.91 (95% CI: 0.77-1.00) for fibrosis. The AI models consistently outperformed expert radiologists, whose interobserver agreement was moderate (κ = 0.58-0.65). CONCLUSION: This study demonstrates that deep learning models can accurately identify steatosis and fibrosis from ultrasound images across diverse populations and clinical settings.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Postgraduate Institute of Medical Education and Research, Chandigarh, India.
- Postgraduate Institute of Medical Education and Research, Chandigarh, India. pankajgupta959@gmail.com.
Förderung
- Science Energy Research Board, Department of Science and Technology, Government of India
Zitieren
APA
Bose, K., Mudgil, P., Gupta, P., Ralmilay, S., Dutta, N., Gulati, A., Kalra, N., Premkumar, M., Taneja, S., Verma, N., De, A., & Duseja, A. (2025). Deep learning for non-invasive detection of steatosis and fibrosis in MASLD: a multicenter study with a new fibroscan-labelled ultrasound dataset. Abdominal Radiology, 51(6), 2906-2918. https://doi.org/10.1007/s00261-025-05309-9
Vancouver
Bose K, Mudgil P, Gupta P, Ralmilay S, Dutta N, Gulati A, et al. Deep learning for non-invasive detection of steatosis and fibrosis in MASLD: a multicenter study with a new fibroscan-labelled ultrasound dataset. Abdominal Radiology. 2025;51(6):2906-2918. doi:10.1007/s00261-025-05309-9
BibTeX
@article{Bose2025,
author = {Bose, Kartik and Mudgil, Priya and Gupta, Pankaj and Ralmilay, Samonee and Dutta, Niharika and Gulati, Ajay and Kalra, Naveen and Premkumar, Madhumita and Taneja, Sunil and Verma, Nipun and De, Arka and Duseja, Ajay},
title = {Deep learning for non-invasive detection of steatosis and fibrosis in MASLD: a multicenter study with a new fibroscan-labelled ultrasound dataset},
journal = {Abdominal Radiology},
year = {2025},
volume = {51},
number = {6},
pages = {2906--2918},
doi = {10.1007/s00261-025-05309-9}
}
EUVIMED