Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images
IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control · 2025
DOI: 10.1109/tuffc.2025.3555180 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2025. Verfasst von Pedro Vianna und 7 weiteren Autor:innen. Hauptthemen laut MEDLINE: Fatty Liver, Image Interpretation, Computer-Assisted, Unsupervised Machine Learning. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: Institute of Data Valorization, Canada Institute for Advanced Research (CIFAR) Artificial Intelligence (AI) Chair, Fonds de Recherche du Québec.
Abstract (Originalsprache)
Ultrasound (US) is considered a key modality for the clinical assessment of hepatic steatosis (i.e., fatty liver) due to its noninvasiveness and availability. Deep learning methods have attracted considerable interest in this field, as they are capable of learning patterns in a collection of images and achieve clinically comparable levels of accuracy in steatosis grading. However, variations in patient populations, acquisition protocols, equipment, and operator expertise across clinical sites can introduce domain shifts that reduce model performance when applied outside the original training setting. In response, unsupervised domain adaptation techniques are being investigated to address these shifts, allowing models to generalize more effectively across diverse clinical environments. In this work, we propose a test-time batch normalization (TTN) technique designed to handle domain shift, especially for changes in label distribution, by adapting selected features of batch normalization (BatchNorm) layers in a trained convolutional neural network model. This approach operates in an unsupervised manner, allowing robust adaptation to new distributions without access to label data. The method was evaluated on two abdominal US datasets collected at different institutions, assessing its capability in mitigating domain shift for hepatic steatosis classification. The proposed method reduced the mean absolute error in steatosis grading by 37% and improved the area under the receiver operating characteristic curves (AUC) for steatosis detection from 0.78 to 0.97, compared to nonadapted models. These findings demonstrate the potential of the proposed method to address domain shift in US-based hepatic steatosis diagnosis, minimizing risks associated with deploying trained models in various clinical settings.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- Institute of Data Valorization
- Canada Institute for Advanced Research (CIFAR) Artificial Intelligence (AI) Chair
- Fonds de Recherche du Québec
- Fonds de Recherche du Québec–Santé (FRQS) and the Fondation de l’Association des Radiologistes du Québec (FARQ) through the Clinical Research Scholarship–Senio…
- IVADO Programme de Recherche Fondamentale
- Institut de Valorisation des Données
Zitieren
APA
Vianna, P., Mehrbod, P., Chaudhary, M., Eickenberg, M., Wolf, G., Belilovsky, E., Tang, A., & Cloutier, G. (2025). Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control. https://doi.org/10.1109/tuffc.2025.3555180
Vancouver
Vianna P, Mehrbod P, Chaudhary M, Eickenberg M, Wolf G, Belilovsky E, et al. Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control. 2025. doi:10.1109/tuffc.2025.3555180
BibTeX
@article{Vianna2025,
author = {Vianna, Pedro and Mehrbod, Paria and Chaudhary, Muawiz and Eickenberg, Michael and Wolf, Guy and Belilovsky, Eugene and Tang, An and Cloutier, Guy},
title = {Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images},
journal = {IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control},
year = {2025},
doi = {10.1109/tuffc.2025.3555180}
}
EUVIMED