Generating and Weighting Semantically Consistent Sample Pairs for Ultrasound Contrastive Learning.
IEEE Trans Med Imaging
; 42(5): 1388-1400, 2023 05.
Article
in English
| MEDLINE | ID: covidwho-2322403
ABSTRACT
Well-annotated medical datasets enable deep neural networks (DNNs) to gain strong power in extracting lesion-related features. Building such large and well-designed medical datasets is costly due to the need for high-level expertise. Model pre-training based on ImageNet is a common practice to gain better generalization when the data amount is limited. However, it suffers from the domain gap between natural and medical images. In this work, we pre-train DNNs on ultrasound (US) domains instead of ImageNet to reduce the domain gap in medical US applications. To learn US image representations based on unlabeled US videos, we propose a novel meta-learning-based contrastive learning method, namely Meta Ultrasound Contrastive Learning (Meta-USCL). To tackle the key challenge of obtaining semantically consistent sample pairs for contrastive learning, we present a positive pair generation module along with an automatic sample weighting module based on meta-learning. Experimental results on multiple computer-aided diagnosis (CAD) problems, including pneumonia detection, breast cancer classification, and breast tumor segmentation, show that the proposed self-supervised method reaches state-of-the-art (SOTA). The codes are available at https//github.com/Schuture/Meta-USCL.
Full text:
Available
Collection:
International databases
Database:
MEDLINE
Main subject:
Diagnosis, Computer-Assisted
/
Neural Networks, Computer
Type of study:
Diagnostic study
Language:
English
Journal:
IEEE Trans Med Imaging
Year:
2023
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS