Journal of Computer Applications ›› 2024, Vol. 44 ›› Issue (11): 3617-3622.DOI: 10.11772/j.issn.1001-9081.2023111650

• Multimedia computing and computer simulation • Previous Articles     Next Articles

Polyp segmentation algorithm based on context-aware network

Cong GU(), Qiqiang DUAN, Siyu REN   

  1. School of Mathematics and Information Science,Zhongyuan University of Technology,Zhengzhou Henan 451191,China
  • Received:2023-12-01 Revised:2024-05-20 Accepted:2024-05-21 Online:2024-06-05 Published:2024-11-10
  • Contact: Cong GU
  • About author:DUAN Qiqiang, born in 1999, M. S. candidate. His research interests include medical image processing, deep learning.
    REN Siyu, born in 1999, M. S. candidate. Her research interests include image processing, deep learning.
  • Supported by:
    Humanities and Social Sciences Research Project of Education Department of Henan Province(2022-ZZJH-098);Preponderant Disciplines Strength Improvement Program of Zhongyuan University of Technology(SD202237)

基于上下文感知网络的息肉分割算法

顾聪(), 段其强, 任思雨   

  1. 中原工学院 数学与信息科学学院,郑州 451191
  • 通讯作者: 顾聪
  • 作者简介:段其强(1999—),男,河南濮阳人,硕士研究生,主要研究方向:医学图像处理、深度学习
    任思雨(1999—),女,河南许昌人,硕士研究生,主要研究方向:图像处理、深度学习。
  • 基金资助:
    河南省教育厅人文社会科学研究项目(2022-ZZJH-098);中原工学院优势学科实力提升计划项目(SD202237)

Abstract:

Deep learning-based methods for polyp image segmentation face the following problems: images captured by different medical devices differ in feature distribution, resulting in domain bias between different polyp segmentation datasets; most existing models focus on processing features of the same scale size, and there are some limitations in their abilities to capture polyps of different scales; the visual features and color differences between a polyp and the surrounding tissue are usually small, making it difficult for the model to accurately distinguish the polyp from the background. To solve these problems, a Context-Aware Network (CANet) with Pyramid Vision Transformer (PVT) as the main part was proposed, which mainly contains the following modules: 1) Domain Adaptive Denoising Module (DADM), which employs channel attention and spatial attention to the low-level feature maps to solve the problem of domain bias and noise between images of different domains; 2) Scale Recalibration Module (SRM), which processes multi-scale features extracted by the encoder to solve the problem of the obvious changes in the size and shape of polyps; 3) Iterative Semantic Embedding Module (ISEM), which reduces background interference, improves perception of the target boundary, and enhances the accuracy of polyp segmentation. Experimental results on five publicly available colon polyp datasets show that CANet achieves better results than current widely used colon polyp segmentation methods, with mDice of 92.6% and 94.0% on Kvasir-SEG and CVC?ClinicDB datasets, respectively.

Key words: medical image segmentation, Transformer, polyp segmentation, adaptive denoising, global attention mechanism

摘要:

基于深度学习的方法进行息肉图像分割时会面临以下问题:不同医疗设备采集的图像在特征分布上存在差异,导致不同的息肉分割数据集之间存在域偏移;现有模型大多专注于处理相同尺度大小的特征,对不同尺度息肉的捕捉能力存在一定的限制;息肉与周围组织的视觉特征和颜色差异通常较小,导致模型难以准确地区分息肉与背景。为了解决这些问题,提出以金字塔结构的视觉Transformer(PVT)为主体的上下文感知网络(CANet),主要包括以下模块:1)域自适应去噪模块(DADM),对低级特征图采用通道注意力以及空间注意力以解决不同域图像之间的域偏移以及噪声问题;2)尺度校准模块(SRM),处理编码器提取的多尺度特征,解决息肉大小和形状明显变化的问题;3)迭代式语义嵌入模块(ISEM),减少背景干扰,提高对目标边界的感知能力,提升息肉分割的准确性。在5个公开的结肠息肉数据集上的实验结果表明,CANet比当前广泛采用的结肠息肉分割方法取得了更好的结果,在Kvasir-SEG和CVC-ClinicDB数据集上的mDice(mean Dice)分别为92.6%和94.0%。

关键词: 医学图像分割, Transformer, 息肉分割, 自适应去噪, 全局注意力机制

CLC Number: