PBE-UNet: A light weight Progressive Boundary-Enhanced U-Net with Scale-Aware Aggregation for Ultrasound Image Segmentation

arXiv cs.CV / 4/16/2026

💬 OpinionSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces PBE-UNet, a lightweight progressive boundary-enhanced U-Net designed to improve lesion segmentation in ultrasound images despite low contrast, blurry boundaries, and large scale variations.
It adds a scale-aware aggregation module (SAAM) that dynamically adjusts receptive fields to capture robust multi-scale context, addressing performance drops caused by varying lesion sizes.
It proposes a boundary-guided feature enhancement (BGFE) approach that progressively expands boundary predictions into wider spatial attention maps to cover broader segmentation error regions.
Experiments on four ultrasound benchmarks (BUSI, Dataset B, TN3K, BP) report that PBE-UNet outperforms existing state-of-the-art ultrasound segmentation methods.
The authors provide an open-source implementation at the linked GitHub repository for reproducibility and further research use.

Abstract

Accurate lesion segmentation in ultrasound images is essential for preventive screening and clinical diagnosis, yet remains challenging due to low contrast, blurry boundaries, and significant scale variations. Although existing deep learning-based methods have achieved remarkable performance, these methods still struggle with scale variations and indistinct tumor boundaries. To address these challenges, we propose a progressive boundary enhanced U-Net (PBE-UNet). Specially, we first introduce a scale-aware aggregation module (SAAM) that dynamically adjusts its receptive field to capture robust multi-scale contextual information. Then, we propose a boundary-guided feature enhancement (BGFE) module to enhance the feature representations. We find that there are large gaps between the narrow boundary and the wide segmentation error areas. Unlike existing methods that treat boundaries as static masks, the BGFE module progressively expands the narrow boundary prediction into broader spatial attention maps. Thus, broader spatial attention maps could effectively cover the wider segmentation error regions and enhance the model's focus on these challenging areas. We conduct expensive experiments on four benchmark ultrasound datasets, BUSI, Dataset B, TN3K, and BP. The experimental results how that our proposed PBE-UNet outperforms state-of-the-art ultrasound image segmentation methods. The code is at https://github.com/cruelMouth/PBE-UNet.