Swin-SFTNet : Spatial Feature Expansion and Aggregation Using Swin Transformer for Whole Breast Micro-Mass Segmentation

Swin-SFTNet : Spatial Feature Expansion and Aggregation Using Swin Transformer for Whole Breast Micro-Mass Segmentation
复制标题

DOI：
10.1109/isbi53787.2023.10230342
复制
发表时间：
2022-11
期刊：
2023 IEEE 20th International Symposium on Biomedical Imaging (ISBI)
影响因子：
--
通讯作者：
Sharif Amit Kamran;Khondker Fariha Hossain;A. Tavakkoli;G. Bebis;Salah A. Baker
Sharif Amit Kamran;Khondker Fariha Hossain;A. Tavakkoli;G. Bebis;Salah A. Baker
中科院分区：
其他
文献类型：
--
作者：
Sharif Amit Kamran;Khondker Fariha Hossain;A. Tavakkoli;G. Bebis;Salah A. Baker

文献摘要

相似文献

将不同的肿块形状和大小结合到深度学习结构的训练中，使得乳房肿块分割具有挑战性。此外，人工分割不规则形状的块状物体既耗时又容易出错。虽然深度神经网络在乳腺肿块分割中表现出了很好的性能，但对于微小肿块的分割效果不佳。在本文中，我们提出了一种新的基于U网状变压器的架构，称为Swin-SFTNet，它在基于乳房X光摄影的微肿块分割中优于最先进的架构。首先，为了捕捉全局背景，我们设计了一种新的空间特征扩展和聚合块(SFEA)，将连续的线性块转换为结构化的空间特征。然后，将其与SWIN变换块提取的局部线性特征相结合，提高整体精度。我们还引入了一种新的嵌入损失，该损失计算编码块和解码块的线性特征嵌入之间的相似度。该方法在CBIS-DDSM、InBreast和CBIS预训练模型上的分割精度分别提高了3.10%、3.81%和3.13%。
Incorporating various mass shapes and sizes in training deep learning architectures has made breast mass segmentation challenging. Moreover, manual segmentation of masses of irregular shapes is time-consuming and error-prone. Though Deep Neural Network has shown outstanding performance in breast mass segmentation, it fails in segmenting micro-masses. In this paper, we propose a novel U-net-shaped transformer-based architecture, called Swin-SFTNet, that outperforms state-of-the-art architectures in breast mammography-based micro-mass segmentation. Firstly to capture the global context, we designed a novel Spatial Feature Expansion and Aggregation Block(SFEA) that transforms sequential linear patches into a structured spatial feature. Next, we combine it with the local linear features extracted by the swin transformer block to improve overall accuracy. We also incorporate a novel embedding loss that calculates similarities between linear feature embeddings of the encoder and decoder blocks. With this approach, we achieve higher segmentation dice over the state-of-the-art by 3.10% on CBIS-DDSM, 3.81% on InBreast, and 3.13% on CBIS pre-trained model on the InBreast test data set.