Mixed 2D and 3D convolutional network with multi-scale context for lesion segmentation in breast DCE-MRI

•M2D3D-MC focuses on lesion segmentation with a limited number of DCE-MRI slices.•2D and 3D operations are deeply integrated in an end-to-end manner.•Multi-scale block is inserted into U-Net to obtain multi-scale information.•The spatial–temporal features of DCE-MRI is learned with the multi-channel...

Full description

Saved in:
Bibliographic Details
Published in:Biomedical signal processing and control Vol. 68; p. 102607
Main Authors: Wang, Hongyu, Cao, Jiaqi, Feng, Jun, Xie, Yilin, Yang, Di, Chen, Baoying
Format: Journal Article
Language:English
Published: Elsevier Ltd 01.07.2021
Subjects:
Breast DCE-MRI
Convolutional neural network
Lesion segmentation
Medical image analysis
Breast DCE-MRI
Lesion segmentation
Convolutional neural network
Medical image analysis
ISSN:1746-8094, 1746-8108
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•M2D3D-MC focuses on lesion segmentation with a limited number of DCE-MRI slices.•2D and 3D operations are deeply integrated in an end-to-end manner.•Multi-scale block is inserted into U-Net to obtain multi-scale information.•The spatial–temporal features of DCE-MRI is learned with the multi-channel image.•The segmentation performance of M2D3D-MC is better than other methods Breast lesion segmentation in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) is an essential step for breast cancer analysis. 2D networks have the advantages of low inference time and greater transfer-ability while 3D networks are superior in learning the contextual information in volumetric data. However, the performance of the 3D network is severely affected when MRI has a low-resolution ratio in the third dimension. In order to integrate the advantages of 2D and 3D networks, we propose a mixed 2D and 3D convolutional network with multi-scale context (M2D3D-MC) for lesion segmentation in breast DCE-MRI with a limited number of axial slices in the scans. Through serial 2D and 3D convolution, the mixed 2D and 3D convolution module has the ability to exploit the contexts between adjacent slices. And considering the diversity of shape and size for breast lesions, we introduce a multi-scale context extractor block consisting of atrous convolutions with different sampling rates to extract multi-scale image features. We justify the proposed method through extensive experiments on 90 MRI studies. Compared with both 2D and 3D networks, M2D3D-MC achieves the best performance with DSC, SEN, and PPV of 76.4%, 75.9%, and 82.4% respectively. A new paradigm is provided for breast lesion segmentation by combining 2D and 3D convolutions to exploit the contextual information. It demonstrates stronger performance in mixed 2D and 3D model given the limited number of axial slices. Our investigation also reveals that the multi-scale context block is effective for breast lesion segmentation.
ISSN:1746-8094
1746-8108
DOI:10.1016/j.bspc.2021.102607