Weakly-supervised positional contrastive learning: application to cirrhosis classification - Télécom Paris
Communication Dans Un Congrès Année : 2023

Weakly-supervised positional contrastive learning: application to cirrhosis classification

Résumé

Large medical imaging datasets can be cheaply and quickly annotated with low-confidence, weak labels (e.g., radiological scores). Access to high-confidence labels, such as histology-based diagnoses, is rare and costly. Pretraining strategies, like contrastive learning (CL) methods, can leverage unlabeled or weakly-annotated datasets. These methods typically require large batch sizes, which poses a difficulty in the case of large 3D images at full resolution, due to limited GPU memory. Nevertheless, volumetric positional information about the spatial context of each 2D slice can be very important for some medical applications. In this work, we propose an efficient weakly-supervised positional (WSP) contrastive learning strategy where we integrate both the spatial context of each 2D slice and a weak label via a generic kernel-based loss function. We illustrate our method on cirrhosis prediction using a large volume of weakly-labeled images, namely radiological low-confidence annotations, and small strongly-labeled (i.e., high-confidence) datasets. The proposed model improves the classification AUC by 5% with respect to a baseline model on our internal dataset, and by 26% on the public LIHC dataset from the Cancer Genome Atlas. The code is available at: https://github.com/Guerbet-AI/wsp-contrastive.
Fichier principal
Vignette du fichier
miccai23_camera_ready.pdf (1.68 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04157998 , version 1 (18-07-2023)
hal-04157998 , version 2 (19-09-2023)

Identifiants

  • HAL Id : hal-04157998 , version 2

Citer

Emma Sarfati, Alexandre Bône, Marc-Michel Rohé, Pietro Gori, Isabelle Bloch. Weakly-supervised positional contrastive learning: application to cirrhosis classification. 26th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2023,, Oct 2023, Vancouver, Canada. ⟨hal-04157998v2⟩
214 Consultations
44 Téléchargements

Partager

More