Abstract:Lesion segmentation in ultrasound images is an essential yet challenging step for early evaluation and diagnosis of cancers. In recent years, many automatic CNN-based methods have been proposed to assist this task. However, most modern approaches often lack capturing long-range dependencies and prior information making it difficult to identify the lesions with unfixed shapes, sizes, locations, and textures. To address this, we present a novel lesion segmentation framework that guides the model to learn the global information about lesion characteristics and invariant features (e.g., morphological features) of lesions to improve the segmentation in ultrasound images. Specifically, the segmentation model is guided to learn the characteristics of lesions from the global maps using an adversarial learning scheme with a self-attention-based discriminator. We argue that under such a lesion characteristics-based guidance mechanism, the segmentation model gets more clues about the boundaries, shapes, sizes, and positions of lesions and can produce reliable predictions. In addition, as ultrasound lesions have different textures, we embed this prior knowledge into a novel region-invariant loss to constrain the model to focus on invariant features for robust segmentation. We demonstrate our method on one in-house breast ultrasound (BUS) dataset and two public datasets (i.e., breast lesion (BUS B) and thyroid nodule from TNSCUI2020). Experimental results show that our method is specifically suitable for lesion segmentation in ultrasound images and can outperform the state-of-the-art approaches with Dice of 0.931, 0.906, and 0.876, respectively. The proposed method demonstrates that it can provide more important information about the characteristics of lesions for lesion segmentation in ultrasound images, especially for lesions with irregular shapes and small sizes. It can assist the current lesion segmentation models to better suit clinical needs.

Dual-Branch Fully Convolutional Segment Anything Model for Lesion Segmentation in Endoscopic Images

Attention-Guided Network with Densely Connected Convolution for Skin Lesion Segmentation

LeSAM: Adapt Segment Anything Model for Medical Lesion Segmentation

Lesion-Decoupling-Based Segmentation with Large-Scale Colon and Esophageal Datasets for Early Cancer Diagnosis

Lesion Segmentation Framework Based on Convolutional Neural Networks with Dual Attention Mechanism

An efficient framework for lesion segmentation in ultrasound images using global adversarial learning and region-invariant loss

Scale-aware Super-resolution Network with Dual Affinity Learning for Lesion Segmentation from Medical Images

Dual-guided network for endoscopic image segmentation with region and boundary cues

CM-SegNet: A Deep Learning-Based Automatic Segmentation Approach for Medical Images by Combining Convolution and Multilayer Perceptron

Region-to-boundary deep learning model with multi-scale feature fusion for medical image segmentation

Customized Segment Anything Model for Medical Image Segmentation

A Spatio-Temporal Fully Convolutional Network For Breast Lesion Segmentation In Dce-Mri

Mask-Enhanced Segment Anything Model for Tumor Lesion Semantic Segmentation

Weakly Supervised Lesion Co-Segmentation on Ct Scans

Image Segmentation of Endoscopic Cataract Based on Semantic Segmentation

A Diversified Supervised Based U-shape Colorectal Lesion Segmentor with Meaningful Feature Supplement and Multi-Level Residual Attention Mechanism

Boosting medical image segmentation via conditional-synergistic convolution and lesion decoupling

Dr-SAM: U-Shape Structure Segment Anything Model for Generalizable Medical Image Segmentation

SamDSK: Combining Segment Anything Model with Domain-Specific Knowledge for Semi-Supervised Learning in Medical Image Segmentation

SDPN: A Slight Dual-Path Network With Local-Global Attention Guided For Medical Image Segmentation

Semi-automatic Skin Lesion Segmentation Via Fully Convolutional Networks