Unsupervised Test-Time Adaptation for Hepatic Steatosis Grading Using Ultrasound B-Mode Images.

Journal: IEEE transactions on ultrasonics, ferroelectrics, and frequency control
PMID:

Abstract

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.

Authors

  • Pedro Vianna
    From the Department of Imaging and Engineering (P.V., S.I.C., C.L.R., L.P.B., E.M., H.L., S.K., M.C., G.C., A.T.), Laboratory of Biorheology and Medical Ultrasonics (P.V., G.C.), and Clinical Laboratory of Image Processing (E.M., A.T.), Centre de Recherche du Centre Hospitalier de l'Université de Montréal (CRCHUM), Montréal, Canada; Institute of Biomedical Engineering (P.V., G.C.) and Department of Computer Science and Operations Research (S.G., I.R., G.W.), Université de Montréal, Montréal, Canada; Departments of Radiology (S.I.C., P.B., C.L.R., L.P.B., Y.H.L., E.A., J.M., T.M.A., K.N.V., J.M.L., J.S.B., A.T.) and Pathology (B.N.N.), Centre Hospitalier de l'Université de Montréal (CHUM), 1058 rue Saint-Denis, Montréal, QC, Canada H2X 3J4; Department of Computer Engineering, École Polytechnique de Montréal, Montréal, Canada (S.K.); Mila-Quebec Artificial Intelligence Institute, Montréal, Canada (S.G., B.T., I.R., E.B., G.W.); and Department of Computer Science and Software Engineering, Concordia University, Montréal, Canada (B.T., E.B.).
  • Paria Mehrbod
  • Muawiz Chaudhary
  • Michael Eickenberg
    Flatiron Institute, New York, NY, United States of America.
  • Guy Wolf
    Department of Mathematics and Statistics, Université de Montréal, Montréal, Quebec, Canada.
  • Eugene Belilovsky
    From the Department of Imaging and Engineering (P.V., S.I.C., C.L.R., L.P.B., E.M., H.L., S.K., M.C., G.C., A.T.), Laboratory of Biorheology and Medical Ultrasonics (P.V., G.C.), and Clinical Laboratory of Image Processing (E.M., A.T.), Centre de Recherche du Centre Hospitalier de l'Université de Montréal (CRCHUM), Montréal, Canada; Institute of Biomedical Engineering (P.V., G.C.) and Department of Computer Science and Operations Research (S.G., I.R., G.W.), Université de Montréal, Montréal, Canada; Departments of Radiology (S.I.C., P.B., C.L.R., L.P.B., Y.H.L., E.A., J.M., T.M.A., K.N.V., J.M.L., J.S.B., A.T.) and Pathology (B.N.N.), Centre Hospitalier de l'Université de Montréal (CHUM), 1058 rue Saint-Denis, Montréal, QC, Canada H2X 3J4; Department of Computer Engineering, École Polytechnique de Montréal, Montréal, Canada (S.K.); Mila-Quebec Artificial Intelligence Institute, Montréal, Canada (S.G., B.T., I.R., E.B., G.W.); and Department of Computer Science and Software Engineering, Concordia University, Montréal, Canada (B.T., E.B.).
  • An Tang
    Centre de recherche du Centre hospitalier de l'Université de Montréal (CRCHUM), Montreal, Canada.
  • Guy Cloutier