| 邢长征,何雨恒,梁浚锋.基于Kolmogorov-Arnold网络平行反向注意网络模型分割结肠镜图像所示息肉[J].中国医学影像技术,2025,41(6):971~975 |
| 基于Kolmogorov-Arnold网络平行反向注意网络模型分割结肠镜图像所示息肉 |
| Parallel reverse enhance attention network module based on Kolmogorov-Arnold networks for segmenting polyps showed on colonoscopy images |
| 投稿时间：2024-12-26 修订日期：2025-06-03 |
| DOI：10.13929/j.issn.1003-3289.2025.06.026 |
| 中文关键词: 结肠息肉 深度学习 图像处理,计算机辅助 结肠镜检查 |
| 英文关键词:colonic polyps deep learning image processing, computer-assisted colonoscopy |
| 基金项目:2024年度辽宁省教育厅基本科研项目(LJ212410147003)。 |
|
| 摘要点击次数: 531 |
| 全文下载次数: 98 |
| 中文摘要: |
| 目的 观察基于Kolmogorov-Arnold网络的平行反向注意网络模型(KAN-PrdaModule)用于分割结肠镜图像中的息肉的价值。方法 选取Kvasir-SEG数据集中的900幅及CVC-ClinicDB数据集中的550幅结肠镜图像为训练集(n=1 450),以ETIS数据集中的196幅、CVC-ClinicDB数据集中的62幅、CVC-ColonDB数据集中的380幅及Kvasir-SEG数据集中的100幅结肠镜图像为测试集(n=738)。基于U-Net改进的KAN-PradModule边缘检测算法、通过多尺度特征融合提升检测精度,采用平均戴斯相似系数(mDSC)、平均交并比(mIoU)、加权度量(Fωβ)及结构度量(Sα),与U-Net、U-Net++、随机前沿生产分析(SFA)及PraNet模型对比,评估上述方法用于分割结肠镜图像所示息肉的价值。结果 5个模型中,SFA模型分割结肠镜图像所示息肉的表现极差,分割息肉边缘模糊且存在部分遗漏;U-Net和U-Net++模型效果尚可,可大致识别息肉;PraNet模型效果较好,分割息肉边缘多清晰;KAN-PrdaModule表现最为出色,与真值图像相似度较高,整体mDSC、mIoU、Fωβ及Sα最优。结论 利用KAN-PrdaMoudel能有效分割结肠镜图像所示息肉,且分割效果优于U-Net、U-Net++、SAF及PraNet模型。 |
| 英文摘要: |
| Objective To observe the value of parallel reverse enhance attention network module based on Kolmogorov-Arnold networks (KAN-PrdaModule) for segmenting polyps showed on colonoscopy images. Methods Nine hundred colonoscopy images in Kvasir-SEG dataset and 550 colonoscopy images in CVC-ClinicDB dataset were selected as training set (n=1 450), while 196 colonoscopy images in ETIS dataset, 62 in CVC-ClinicDB dataset, 380 in CVC-ColonDB dataset and 100 in Kvasir-SEG dataset were enrolled as test set (n=738). KAN-PrdaModule was proposed through improving U-Net, which improved detection accuracy through multi-scale feature fusion, and the value of KAN-PrdaModule for segmenting polyps showed on colonoscopy images was analyzed according to mean Dice similarity coefficient (mDSC), mean intersection over union (mIoU), weighted metric (Fωβ) and structural metric (Sα), and compared with U-Net, U-Net++, stochastic frontier analysis (SFA) and PraNet models. Results Among the above 5 models, the performance of SFA model for segmenting polyps on colonoscopy images was poor, with blurry edges of polyps and some ones were missed. U-Net and U-Net++ models had decent performance, which could roughly identify polyps. PraNet model performed well, and the segmented edges of polyps were clear. KAN-PrdaModule had the best performance, showed high similarity to the true value images, with the best overall mDSC, mIoU, Fωβ and Sα. Conclusion KAN-PrdaModule could effectively segment polyps showed on colonoscopy images, with segmenting effect better than U-Net, U-Net++, SFA and PraNet models. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
