Self Supervised Lesion Recognition For Breast Ultrasound Diagnosis

Yuanfan Guo,Canqian Yang,Tiancheng Lin,Chunxiao Li,Rui Zhang,Yi Xu

DOI: https://doi.org/10.48550/arXiv.2204.08477

2022-04-19

Abstract:Previous deep learning based Computer Aided Diagnosis (CAD) system treats multiple views of the same lesion as independent images. Since an ultrasound image only describes a partial 2D projection of a 3D lesion, such paradigm ignores the semantic relationship between different views of a lesion, which is inconsistent with the traditional diagnosis where sonographers analyze a lesion from at least two views. In this paper, we propose a multi-task framework that complements Benign/Malignant classification task with lesion recognition (LR) which helps leveraging relationship among multiple views of a single lesion to learn a complete representation of the lesion. To be specific, LR task employs contrastive learning to encourage representation that pulls multiple views of the same lesion and repels those of different lesions. The task therefore facilitates a representation that is not only invariant to the view change of the lesion, but also capturing fine-grained features to distinguish between different lesions. Experiments show that the proposed multi-task framework boosts the performance of Benign/Malignant classification as two sub-tasks complement each other and enhance the learned representation of ultrasound images.

Image and Video Processing,Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The problem that this paper attempts to solve is that in breast ultrasound diagnosis, the existing computer - aided diagnosis (CAD) systems based on deep learning usually regard the images of different views of the same lesion as independent image processing, ignoring the semantic relationships between the images of different views. This is inconsistent with the method in which doctors diagnose lesions by analyzing images of at least two views in traditional diagnosis. This processing method leads to an incomplete representation of the lesion and affects the accuracy of benign - malignant classification. To solve this problem, the paper proposes a multi - task framework, in which the lesion recognition (LR) task is introduced as a supplement to the benign / malignant classification task. The LR task, through contrastive learning, encourages the model to learn a representation method that can bring the different views of the same lesion closer and push the views of different lesions apart. In this way, the model can not only maintain invariance when changing views, but also capture fine - grained features for distinguishing different lesions, thereby improving the performance of benign - malignant classification. The experimental results show that the proposed multi - task framework enhances the learning representation of ultrasound images through the complementary effect of the two sub - tasks and improves the performance of benign - malignant classification. Specifically, after the introduction of the LR task, the area under the receiver operating characteristic curve (AUC) of the main classification task is increased by 1.5%, and the accuracy is increased by 2.8%. In addition, the LR task also reduces the prediction inconsistency among multiple views of the same lesion, making the diagnosis more accurate.

Self Supervised Lesion Recognition For Breast Ultrasound Diagnosis

Combining Convolutional and Recurrent Neural Networks for Classification of Focal Liver Lesions in Multi-phase CT Images.

Weakly Supervised Lesion Detection and Diagnosis for Breast Cancers with Partially Annotated Ultrasound Images

Multi-label Recognition of Cancer-Related Lesions with Clinical Priors on White-Light Endoscopy

Breast Lesion Diagnosis Using Static Images and Dynamic Video

An experimental study on breast lesion detection and classification from ultrasound images using deep learning architectures

Automatic Detection of Breast Lesions in Automated 3D Breast Ultrasound with Cross-Organ Transfer Learning

A Novel Deep Learning System for Breast Lesion Risk Stratification in Ultrasound Images

Spatial Attention Lesion Detection On Automated Breast Ultrasound

A Novel Deep Learning Model for Breast Tumor Ultrasound Image Classification with Lesion Region Perception

Enhancing AI Diagnostics: Autonomous Lesion Masking via Semi-Supervised Deep Learning

Joint Lesion Detection and Classification of Breast Ultrasound Video Via a Clinical Knowledge-Aware Framework

An efficient framework for lesion segmentation in ultrasound images using global adversarial learning and region-invariant loss

Domain Knowledge Powered Deep Learning for Breast Cancer Diagnosis Based on Contrast-Enhanced Ultrasound Videos

Multi-task Learning for Segmentation and Classification of Tumors in 3D Automated Breast Ultrasound Images

BIRADS Features-Oriented Semi-supervised Deep Learning for Breast Ultrasound Computer-Aided Diagnosis

Universal Lesion Detection by Learning from Multiple Heterogeneously Labeled Datasets

Segmentation Information with Attention Integration for Classification of Breast Tumor in Ultrasound Image

Breast Tumor Detection In Ultrasound Images Using Deep Learning

Multimodal Ultrasound Imaging Based Diagnosis of Liver Cancers with a Two-Stage Multi-View Learning Framework.

Attention-Guided Learning with Feature Reconstruction for Skin Lesion Diagnosis using Clinical and Ultrasound Images