A deep neural network improves endoscopic detection of early gastric cancer without blind spots
A deep neural network improves endoscopic detection of early gastric cancer without blind spots
复制标题
DOI：
10.1055/a-0855-3532
复制
发表时间：
2019-06-01
期刊：
影响因子：
9.3
通讯作者：
Yu, Honggang
中科院分区：
文献类型：
--
作者：
Wu, Lianlian;Zhou, Wei;Yu, Honggang
背景胃癌是全球第三大致死性恶性肿瘤。最近开发了一种用于执行视觉任务的新型深度卷积神经网络（DCNN）。目的：建立一个基于DCNN的无盲点早期胃癌（EGC）检测系统。方法收集3170例胃癌和5981例良性病变图像，训练DCNN进行EGC检测。共收集了24549张来自胃不同部位的图像，以训练DCNN来监测盲点。开发了类激活图以自动覆盖可疑的癌区域。结果DCNN诊断EGC的准确率为92.5%，敏感性为94.0%，特异性为91.0%，阳性预测值为91.3%，阴性预测值为93.8%，优于所有级别的内镜医师。在将胃位置分类为10或26个部分的任务中，DCNN实现了90%或65.9%的准确率，与专家的表现相当。在实时未处理的EGD视频中，DCNN实现了检测EGC和监测盲点的自动化性能。结论我们开发了一种基于DCNN的系统，可以准确检测EGC，并比内窥镜医生更好地识别胃位置，并主动跟踪可疑的癌性病变，并在EGD期间监测盲点。
Background Gastric cancer is the third most lethal malignancy worldwide. A novel deep convolution neural network (DCNN) to perform visual tasks has been recently developed. The aim of this study was to build a system using the DCNN to detect early gastric cancer (EGC) without blind spots during esophagogastroduodenoscopy (EGD).Methods 3170 gastric cancer and 5981 benign images were collected to train the DCNN to detect EGC. A total of 24549 images from different parts of stomach were collected to train the DCNN to monitor blind spots. Class activation maps were developed to automatically cover suspicious cancerous regions. A grid model for the stomach was used to indicate the existence of blind spots in unprocessed EGD videos.Results The DCNN identified EGC from non-malignancy with an accuracy of 92.5%, a sensitivity of 94.0%, a specificity of 91.0%, a positive predictive value of 91.3%, and a negative predictive value of 93.8%, outperforming all levels of endoscopists. In the task of classifying gastric locations into 10 or 26 parts, the DCNN achieved an accuracy of 90% or 65.9%, on a par with the performance of experts. In realtime unprocessed EGD videos, the DCNN achieved automated performance for detecting EGC and monitoring blind spots.Conclusions We developed a system based on a DCNN to accurately detect EGC and recognize gastric locations better than endoscopists, and proactively track suspicious cancerous lesions and monitor blind spots during EGD.