Integrating Plain 2D Ultrasound and Clinical Data for Identification of At‐Risk MASH: A Multicenter Proof of Concept Study
Journal of Gastroenterology and Hepatology · 16.10.2026
DOI: 10.1111/jgh.70763 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Gastroenterology and Hepatology, online erschienen am 16.10.2026. Verfasst von Vigneshwaran Selvakumar und 21 weiteren Autor:innen. Beteiligte Einrichtung: Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 37 Literaturangaben.
Abstract (Originalsprache)
Background
Identifying people with at‐risk metabolic dysfunction‐associated steatohepatitis (MASH) is a major challenge. We aimed to use deep learning to develop a convolutional neural network (CNN) model to facilitate the detection of at‐risk MASH.
Methods
A multi‐center cohort study was conducted across five tertiary hospitals in Japan and China. Clinical data and conventional ultrasound images were retrospectively collected. Organ segmentation using YOLO version 11 with quality control was performed, followed by creation of an EfficientNet‐B6 CNN regression model, with image‐level predictions averaged per patient, to estimate FAST and AGILE 3+ risk categories. Model performance was assessed using agreement across clinically relevant rule‐out and rule‐in thresholds, R 2 , root mean squared error (RMSE), and area under the receiver operating curve (AUROC).
Results
The final dataset comprised 416 patients for training (7643 images), 139 for internal validation (2463 images), and 154 for external testing (3539 images). The CNN model demonstrated good agreement with FAST ( R 2 = 0.684; RMSE = 0.131; category agreement = 0.71) and moderate agreement with AGILE 3+ ( R 2 = 0.445; RMSE = 0.1964; category agreement = 0.63). For FAST, AUROC was 0.974 (95% CI 0.947–0.993) for 0.35 and 0.932 (95% CI 0.884–0.969) for the 0.67 cut‐off. For AGILE 3+, AUROC was 0.908 (95% CI 0.854–0.950) for 0.451 and 0.902 (95% CI 0.839–0.953) for the 0.679 cut‐off.
Conclusion
The CNN model applied to clinical ultrasound images can approximate the accuracy of VCTE‐based composite risk scores for detecting at‐risk MASH. This may support the detection of at‐risk MASH in settings with limited access to transient elastography.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore.
- Yong Loo Lin School of Medicine, National University of Singapore, Singapore.
- Department of Diagnostic Imaging, National University Hospital, Singapore, Singapore.
- Department of Ultrasound, First Affiliated Hospital of Wenzhou Medical University, Wenzhou, China.
- Division of Gastroenterology, Department of Medicine, Kurume University School of Medicine, Kurume, Japan.
- Department of Gastroenterology and Hepatology, Yokohama City University School of Medicine Graduate School of Medicine, Yokohama, Kanagawa, Japan.
ORCID-Profile: Jordan Jun Yi Low, Nicole S Tang, Chuan X Lau, Kalpenya D Ganakirthana, Shun-Ping Chen, Masahito Nakano, Keisuke Amano, Asako Nogami, Satoshi Oeda, Masato Yoneda, Yuya Seko, Minghua Zheng, Takumi Kawaguchi, Hirokazu Takahashi, Cheng Han Ng, Nicholas Syn, Mei Chin Lim, Margaret L P Teng
Zitieren
APA
Selvakumar, V., Low, J. J. Y., Tang, N. S., Lau, C. X., Ganakirthana, K. D., Li, Z., Xia, Y. S., Chen, S. P., Nakano, M., Amano, K., Nogami, A., Oeda, S., Yoneda, M., Seko, Y., Zheng, M., Kawaguchi, T., Takahashi, H., Ng, C. H., Syn, N., …, & Huang, D. Q. (2026). Integrating Plain 2D Ultrasound and Clinical Data for Identification of At‐Risk MASH: A Multicenter Proof of Concept Study. Journal of Gastroenterology and Hepatology. https://doi.org/10.1111/jgh.70763
Vancouver
Selvakumar V, Low JJY, Tang NS, Lau CX, Ganakirthana KD, Li Z, et al. Integrating Plain 2D Ultrasound and Clinical Data for Identification of At‐Risk MASH: A Multicenter Proof of Concept Study. Journal of Gastroenterology and Hepatology. 2026. doi:10.1111/jgh.70763
BibTeX
@article{Selvakumar2026,
author = {Selvakumar, Vigneshwaran and Low, Jordan Jun Yi and Tang, Nicole S and Lau, Chuan X and Ganakirthana, Kalpenya D and Li, Zongchen and Xia, Yong-Shen and Chen, Shun-Ping and Nakano, Masahito and Amano, Keisuke and Nogami, Asako and Oeda, Satoshi and Yoneda, Masato and Seko, Yuya and Zheng, Minghua and Kawaguchi, Takumi and Takahashi, Hirokazu and Ng, Cheng Han and Syn, Nicholas and Lim, Mei Chin and Teng, Margaret L P and Huang, Daniel Q},
title = {Integrating Plain 2D Ultrasound and Clinical Data for Identification of At‐Risk MASH: A Multicenter Proof of Concept Study},
journal = {Journal of Gastroenterology and Hepatology},
year = {2026},
doi = {10.1111/jgh.70763}
}
EUVIMED