Colorectal Polyp Classification Based On Latent Sharing Features Domain from Multiple Endoscopy Images
Colorectal Polyp Classification Based On Latent Sharing Features Domain from Multiple Endoscopy Images
复制标题
基于多个内窥镜图像的潜在共享特征域的结直肠息肉分类
DOI：
10.1016/j.procs.2020.09.325
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Kunio Kasugai
中科院分区：
文献类型：
--
作者：
Hiroyasu Usami;Yuji Iwahori;Yoshinori Adachi;M. K. Bhuyan;Aili Wang;Satoshi Inoue;Masahide Ebi;Naotaka Ogasawara;Kunio Kasugai
作为一种从内窥镜图像判断息肉良恶性的方法，已经提出了一些使用超高倍内窥镜的方法。超高放大倍率内窥镜使诊断在细胞水平。然而，它往往花费很多时间进行诊断，并需要特定的昂贵设备。常规内窥镜的诊断有三种类型的图像：白光、染料和一般的窄带图像（NBI）。本文提出了一种利用常规内窥镜拍摄的息肉图像进行良、恶性分类的方法。通过将预训练好的CNN适应于每个域，提取内窥镜图像的每个图像特征。最后，利用提取的特征对息肉进行分类。实验证实，该方法对结直肠息肉的良恶性分类准确率超过90%。
As a method to judge the benign or malignant polyp from endoscope images, some methods have been proposed using an ultra-high magnification endoscope. The ultra-high magnification endoscope enables the diagnosis at the cell level. However, it tends to spend many times for diagnosis and requires specific expensive devices. There are three types of images that are taken for diagnosis by the regular endoscope: white light, dye, and narrowband image (NBI) in general. This paper proposes a benign/malignant polyp classification method using these images taken by the regular endoscope. Each image features derived from endoscope images are extracted by adapting a pre-trained CNN to each domain. Finally, polyps are classified using extracted features. Experiments confirmed that the proposed method enabled the classification of benign or malignant colorectal polyps with over 90% accuracy.