Abstract:Background: The segmentation of breast ultrasound (US) images has been a challenging task, mainly due to limited data and the inherent image characteristics involved, such as low contrast and speckle noise. Although convolutional neural network-based (CNN-based) methods have made significant progress over the past decade, they lack the ability to model long-range interactions. Recently, the transformer method has been successfully applied to the tasks of computer vision. It has a strong ability to capture distant interactions. However, most transformer-based methods with excellent performance rely on pre-training on large datasets, making it infeasible to directly apply them to medical images analysis, especially that of breast US images with limited high-quality labels. Therefore, it is of great significance to find a robust and efficient transformer-based method for use on small breast US image datasets.Methods: We developed a dilated transformer (DT) method which mainly uses the proposed residual axial attention layers to build encoder blocks and the introduced dilation module (DM) to further increase the receptive field. We evaluated the proposed method on 2 breast US image datasets using the 5-fold cross-validation method. Dataset A was a public dataset with 562 images, while dataset B was a private dataset with 878 images. Ground truth (GT) was delineated by 2 radiologists with more than 5 years of experience. The evaluation was followed by related ablation experiments.Results: The DT was found to be comparable with the state-of-the-art (SOTA) CNN-based method and outperformed the related transformer-based method, medical transformer (MT), on both datasets. Especially on dataset B, the DT outperformed the MT on metrics of Jaccard index (JI) and Dice similarity coefficient (DSC) by 2.67% and 4.68%, respectively. Meanwhile, when compared with Unet, the DT improved JI and DSC by 4.89% and 4.66%, respectively. Moreover, the results of the ablation experiments showed that each add-on part of the DT is important and contributes to the segmentation accuracy.Conclusions: The proposed transformer-based method could achieve advanced segmentation performance on different small breast US image datasets without pretraining.

A dual-stage transformer and MLP-based network for breast ultrasound image segmentation

A Multi-Task Transformer with Local-Global Feature Interaction and Multiple Tumoral Region Guidance for Breast Cancer Diagnosis

HAU-Net: Hybrid CNN-transformer for breast ultrasound image segmentation

A dual-branch and dual attention transformer and CNN hybrid network for ultrasound image segmentation

A Hybrid Enhanced Attention Transformer Network for Medical Ultrasound Image Segmentation

A Multiscale Nonlocal Feature Extraction Network for Breast Lesion Segmentation in Ultrasound Images.

CSwin-PNet: A CNN-Swin Transformer combined pyramid network for breast lesion segmentation in ultrasound images

MF-Net: Multiple-feature Extraction Network for Breast Lesion Segmentation in Ultrasound Images

HCTNet: A hybrid CNN-transformer network for breast ultrasound image segmentation

Swin-Net: A Swin-Transformer-Based Network Combing with Multi-Scale Features for Segmentation of Breast Tumor Ultrasound Images

MFMSNet: A Multi-frequency and Multi-scale Interactive CNN-Transformer Hybrid Network for breast ultrasound image segmentation

Dilated transformer: residual axial attention for breast ultrasound image segmentation

Multi-task approach based on combined CNN-transformer for efficient segmentation and classification of breast tumors in ultrasound images

U-MLP: MLP-based ultralight refinement network for medical image segmentation

Transfer Learning for Automatic Joint Segmentation of Thyroid and Breast Lesions from Ultrasound Images.

Multi-dimension unified Swin Transformer for 3D Lesion Segmentation in Multiple Anatomical Locations

Automatic tumor segmentation in breast ultrasound images using a dilated fully convolutional network combined with an active contour model.

SUTrans-NET: a hybrid transformer approach to skin lesion segmentation

MT-TransUNet: Mediating Multi-Task Tokens in Transformers for Skin Lesion Segmentation and Classification

Ultrasonic breast tumor extraction based on adversarial mechanism and active contour

Multi-Level Swin Transformer Enabled Automatic Segmentation and Classification of Breast Metastases