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

Automatic segmentation of breast lesions from ultrasound images plays an important role in computer-aided breast cancer diagnosis. Many deep learning methods based on convo-lutional neural networks (CNNs) have been proposed for breast ultrasound image segmen-tation. However, breast ultrasound image segmentation is still challenging due to ambiguous lesion boundaries. We propose a novel dual-stage framework based on Trans-former and Multi-layer perceptron (MLP) for the segmentation of breast lesions. We com-bine the Swin Transformer block with an efficient pyramid squeezed attention block in a parallel design and introduce bi-directional interactions across branches, which can effi-ciently extract multi-scale long-range dependencies to improve the segmentation perfor-mance and robustness of the model. Furthermore, we introduce tokenized MLP block in the MLP stage to extract global contextual information while retaining fine-grained infor-mation to segment more complex breast lesions. We have conducted extensive experi-ments with state-of-the-art methods on three breast ultrasound datasets, including BUSI, BUL, and MT_BUS datasets. The dice coefficient reached 0.8127 +/- 0.2178, and the intersection over union reached 0.7269 +/- 0.2370 on benign lesions when the Hausdorff dis-tance was maintained at 3.75 +/- 1.83. The dice coefficient of malignant lesions is improved by 3.09% for BUSI dataset. The segmentation results on the BUL and MT_BUS datasets also show that our proposed model achieves better segmentation results than other methods. Moreover, the external experiments indicate that the proposed model provides better gen-eralization capability for breast lesion segmentation. The dual-stage scheme and the pro-posed Transformer module achieve the fine-grained local information and long-range dependencies to relieve the burden of radiologists.(c) 2023 Published by Elsevier B.V. on behalf of Nalecz Institute of Biocybernetics and Bio-medical Engineering of the Polish Academy of Sciences.