Department of Anesthesia Surgery, The Affiliated Huaian No.1 People's Hospital of Nanjing Medical University, Huai'an, 223300, China.
Department of Cardiothoracic Surgery, The Affiliated Huaian No.1 People's Hospital of Nanjing Medical University, Huai'an, 223300, China.
Sci Rep. 2024 Oct 5;14(1):23179. doi: 10.1038/s41598-024-72763-8.
Colonoscopy is widely recognized as the most effective method for the detection of colon polyps, which is crucial for early screening of colorectal cancer. Polyp identification and segmentation in colonoscopy images require specialized medical knowledge and are often labor-intensive and expensive. Deep learning provides an intelligent and efficient approach for polyp segmentation. However, the variability in polyp size and the heterogeneity of polyp boundaries and interiors pose challenges for accurate segmentation. Currently, Transformer-based methods have become a mainstream trend for polyp segmentation. However, these methods tend to overlook local details due to the inherent characteristics of Transformer, leading to inferior results. Moreover, the computational burden brought by self-attention mechanisms hinders the practical application of these models. To address these issues, we propose a novel CNN-Transformer hybrid model for polyp segmentation (CTHP). CTHP combines the strengths of CNN, which excels at modeling local information, and Transformer, which excels at modeling global semantics, to enhance segmentation accuracy. We transform the self-attention computation over the entire feature map into the width and height directions, significantly improving computational efficiency. Additionally, we design a new information propagation module and introduce additional positional bias coefficients during the attention computation process, which reduces the dispersal of information introduced by deep and mixed feature fusion in the Transformer. Extensive experimental results demonstrate that our proposed model achieves state-of-the-art performance on multiple benchmark datasets for polyp segmentation. Furthermore, cross-domain generalization experiments show that our model exhibits excellent generalization performance.
结肠镜检查被广泛认为是检测结肠息肉的最有效方法，这对于结直肠癌的早期筛查至关重要。在结肠镜图像中识别和分割息肉需要专业的医学知识，而且通常是劳动密集型和昂贵的。深度学习为息肉分割提供了一种智能和高效的方法。然而，息肉大小的可变性以及息肉边界和内部的异质性给准确分割带来了挑战。目前，基于 Transformer 的方法已成为息肉分割的主流趋势。然而，由于 Transformer 的固有特性，这些方法往往忽略了局部细节，导致结果不佳。此外，自注意力机制带来的计算负担阻碍了这些模型的实际应用。为了解决这些问题，我们提出了一种用于息肉分割的新型 CNN-Transformer 混合模型（CTHP）。CTHP 结合了 CNN 在建模局部信息方面的优势和 Transformer 在建模全局语义方面的优势，以提高分割准确性。我们将整个特征图上的自注意力计算转换为宽度和高度方向，显著提高了计算效率。此外，我们设计了一个新的信息传播模块，并在注意力计算过程中引入了附加的位置偏差系数，这减少了 Transformer 中深度和混合特征融合引入的信息分散。广泛的实验结果表明，我们提出的模型在多个息肉分割基准数据集上取得了最先进的性能。此外，跨域泛化实验表明，我们的模型表现出出色的泛化性能。
