基于阶梯结构的U-Net结肠息肉分割算法

Polyp Segmentation Using Stair-structured U-Net

结肠息肉的精确分割对结直肠癌的诊断和治疗具有重要意义,目前的分割方法普遍存在有伪影、分割精度低等问题。该文提出一种基于阶梯结构的U-Net结肠息肉分割算法(SU-Net),使用U-Net的U型结构,利用Kronecker乘积来扩展标准空洞卷积核,构成Kronecker空洞卷积下采样有效扩大感受野,弥补传统空洞卷积容易丢失的细节特征;应用具有阶梯结构的融合模块,遵循扩展和堆叠原则形成阶梯状的分层结构,有效捕获上下文信息并从多个尺度聚合特征;在解码器引入卷积重构上采样模块生成密集的像素级预测图,捕获双线性插值上采样中缺少的精细信息。在Kvasir-SEG数据集和CVC-EndoSceneStill数据集上对模型进行了测试,相似系数(Dice)指标和交并比(IoU)指标分别达到了87.51%,88.75%和82.30%,85.64%。实验结果表明,该文所提方法改善了因过度曝光、低对比度引起的分割精度低的问题,同时消除了边界外部的图像伪影和图像内部不连贯的现象,优于其他息肉分割方法。

The precise segmentation of colon polyps plays a significant role in the diagnosis and treatment of colorectal cancer.The existing segmentation methods have generally artifacts and low segmentation accuracy.In this paper,Stair-structured U-Net(SU-Net)is proposed to segment polyp,using U-shaped structure.The Kronecker product is used to extend the standard atrous convolution kernel to keep more detail structrural features that are easily ignored.Stair-structured fusion module is applied to encompass effectively multi-scale features.The decoder introduces a convolutional reshaped upsampling module to generate pixel-level predictions.Experiments are performed on the Kvasir-SEG dataset and the CVC-EndoSceneStill dataset.The results show that the method proposed in this paper outperforms other polyp segmentation methods in Dice and Intersection-over-Union(IoU).