基于超声内镜的智能胰腺癌变检测网络  

Intelligent Pancreatic Cancer Detection Network Based on Endoscopic Ultrasonography

作　　者：文晓媚 黄丹平 胡珊珊[2] 宁波[3] WEN Xiao-mei;HUANG Dan-ping;HU Shan-shan;NING Bo(College of Mechanical Engineering,Sichuan University of Science&Engineering,Yibin 644000,China;Sichuan Provincial People's Hospital,Chengdu 610000,China;The Second Affiliated Hospital of Chongqing Medical University,Chongqing 400000,China)

机构地区：[1]四川轻化工大学机械工程学院,宜宾644000 [2]四川省人民医院,成都610000 [3]重庆医科大学附属第二医院,重庆400000

出　　处：《科学技术与工程》2022年第34期15203-15212,共10页Science Technology and Engineering

基　　金：四川省重点实验室项目(NJ2018-05);自贡市科技局重点项目(2019YYJC12)。

摘　　要：为对超声内镜检测返回的胰腺图像实现智能癌变诊断,将经典分类识别网络AlexNet和SE注意力机制进行结合,提出的SE-AlexNet网络可以准确检出癌变图像,判别正常图像所属胰腺部位。为验证该算法的可靠性与优胜性,在自制数据集下的一系列比较实验,包括基础网络AlexNet与其他经典分类网络的模型比较实验,以及在不同位置插入不同注意机制改进得到的各种模型的比较实验。结果表明:SE-AlexNet模型总体精确率和召回率可达99.56%和98.69%,对于癌变图像,检测精确率和召回率为100%,能够为现实超声内镜胰腺癌检查实现有效的辅助诊断。To realize intelligent cancer diagnosis for pancreatic images returned by endoscopic ultrasonography,the classic classification and recognition network AlexNet and SE attention mechanism were combined.The proposed SE-AlexNet network can accurately detect cancerous images and discriminate the pancreatic parts to which the normal images belong.In order to verify the reliability and superiority of the algorithm,a series of comparison experiments were carried out under the self-made dataset,including the model comparison experiments between the basic network AlexNet and other classical classification networks,and the comparison experiments of various models improved and obtained by inserting different attention mechanism in different positions.The results show that the overall precision and recall rate of SE-AlexNet model can reach 99.56%and 98.69%.For cancerous images,the detection precision and recall rate are 100%,which can achieve an effective auxiliary diagnosis for practical endoscopic ultrasonography of pancreatic cancer.

关 键 词：超声内镜 胰腺癌 AlexNet 注意力机制 辅助诊断 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]