Real-World Insights in Designing SteatoStat: An End-to-End Deep Learning Pipeline for Hepatic Steatosis Quantification
Diagnostics · 12.06.2026 · 16 (12) · 1825
DOI: 10.3390/diagnostics16121825 ↗Kurz gefasst
Fachartikel in der Zeitschrift Diagnostics, online erschienen am 12.06.2026. Verfasst von Nagalakshmi Jegannathan und 18 weiteren Autor:innen. Beteiligte Einrichtung: Department of Anatomical Pathology, Singapore General Hospital, Singapore 169856, Singapore. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: AI Singapore Tech Challenge Funding Scheme.
Abstract (Originalsprache)
Background
Metabolic dysfunction-associated steatotic liver disease (MASLD) is a significant and escalating global health concern, with an estimated prevalence of 30%. Current assessments of hepatic steatosis, a hallmark of MASLD, rely on semi-quantitative grading by pathologists, which is inherently limited by inter-observer variability.
Objective
To address this limitation, we developed a novel deep learning pipeline, named SteatoStat, to standardize and enhance the quantification of hepatic steatosis in patients with MASLD.
Method
The SteatoStat pipeline employs and integrates multiple components such as file format standardization, rule-based cell filtering, and multiple segmentation models across various liver structures, resulting in an output of a continuous quantitative measure of steatosis percentage and translated into steatosis grades.
Results
We report a high degree of accuracy and reliability with SteatoStat achieving the following performance metrics (DICE score = 0.8955, AUROC = 0.9928, F1 score = 0.8990). When benchmarked against expert pathologists, the weighted Kappa coefficient is 0.837. Furthermore, in comparison with an existing, well-established model, SteatoStat demonstrated a weighted Kappa coefficient = 0.765.
Conclusions
These robust findings underscore its potential clinical utility in providing a standardized objective quantification of hepatic steatosis. Future directions include enhancing the model’s generalizability and its clinical integration through validation on independent, multi-institutional datasets.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Anatomical Pathology, Singapore General Hospital, Singapore 169856, Singapore.
- Institute of High-Performance Computing, Agency for Science, Technology and Research, Singapore 138632, Singapore.
- School of Biological Sciences, Nanyang Technological University, Singapore 637551, Singapore.
- Ministry of Health Holdings, Singapore 139691, Singapore.
- Department of Cardiothoracic and Abdominal Radiology, Singapore General Hospital, Singapore 169608, Singapore.
- Duke-NUS Medical School, Singapore 169857, Singapore.
ORCID-Profile: Nagalakshmi Jegannathan, Jia Xuan Seow, Lionel Tim-Ee Cheng, George Boon Bee Goh, Dean Tai, Chee Leong Cheng, Li Yan Khor, Wei Qiang Leow
Förderung
- AI Singapore Tech Challenge Funding Scheme
Zitieren
APA
Jegannathan, N., Zhang, X., Seow, J. X., Zhou, M., Wang, L., Goh, G. L., Heng, S. Y., Ng, T. R., Goh, R. S. M., Fu, H., Liu, Y., Cheng, L. T., Goh, G. B. B., Tai, D., Cheng, C. L., Wan, W. K., Lim, T. K. H., Khor, L. Y., & Leow, W. Q. (2026). Real-World Insights in Designing SteatoStat: An End-to-End Deep Learning Pipeline for Hepatic Steatosis Quantification. Diagnostics, 16(12), 1825. https://doi.org/10.3390/diagnostics16121825
Vancouver
Jegannathan N, Zhang X, Seow JX, Zhou M, Wang L, Goh GL, et al. Real-World Insights in Designing SteatoStat: An End-to-End Deep Learning Pipeline for Hepatic Steatosis Quantification. Diagnostics. 2026;16(12):1825. doi:10.3390/diagnostics16121825
BibTeX
@article{Jegannathan2026,
author = {Jegannathan, Nagalakshmi and Zhang, Xiaoman and Seow, Jia Xuan and Zhou, Menghan and Wang, Long and Goh, Guo Lin and Heng, Seow Ye and Ng, Tony De Rong and Goh, Rick Siow Mong and Fu, Huazhu and Liu, Yong and Cheng, Lionel Tim-Ee and Goh, George Boon Bee and Tai, Dean and Cheng, Chee Leong and Wan, Wei Keat and Lim, Tony Kiat Hon and Khor, Li Yan and Leow, Wei Qiang},
title = {Real-World Insights in Designing SteatoStat: An End-to-End Deep Learning Pipeline for Hepatic Steatosis Quantification},
journal = {Diagnostics},
year = {2026},
volume = {16},
number = {12},
pages = {1825},
doi = {10.3390/diagnostics16121825}
}
EUVIMED