Masked Video Modeling with Correlation-aware Contrastive Learning for Breast Cancer Diagnosis in Ultrasound

TL;DR

This paper introduces a novel method for breast cancer diagnosis using ultrasound videos, combining masked video modeling and correlation-aware contrastive learning to improve classification accuracy with limited annotations.

Contribution

It presents a pioneering approach that directly utilizes ultrasound videos with a new pretraining technique and a correlation-aware contrastive loss for better lesion differentiation.

Findings

Achieved promising classification performance.

Outperformed other state-of-the-art methods.

Reduced reliance on large annotated datasets.

Abstract

Breast cancer is one of the leading causes of cancer deaths in women. As the primary output of breast screening, breast ultrasound (US) video contains exclusive dynamic information for cancer diagnosis. However, training models for video analysis is non-trivial as it requires a voluminous dataset which is also expensive to annotate. Furthermore, the diagnosis of breast lesion faces unique challenges such as inter-class similarity and intra-class variation. In this paper, we propose a pioneering approach that directly utilizes US videos in computer-aided breast cancer diagnosis. It leverages masked video modeling as pretraining to reduce reliance on dataset size and detailed annotations. Moreover, a correlation-aware contrastive loss is developed to facilitate the identifying of the internal and external relationship between benign and malignant lesions. Experimental results show that our proposed approach achieved promising classification performance and can outperform other state-of-the-art methods.