Wang S X, Ke Y, Liu Y M, Liu S Y, Song S B, He S, Zhang Y M, Dou L Z, Liu Y, Liu X D, Wu H R, Su F X, Zhang F Y, Zhang W, Wang G Q
Department of Endoscopy, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100021, China.
Department of Endoscopy, National Cancer Center/Cancer Hospital& Shenzhen Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Shenzhen 518116, China.
Zhonghua Zhong Liu Za Zhi. 2022 May 23;44(5):395-401. doi: 10.3760/cma.j.cn112152-20211126-00877.
To construct the diagnostic model of superficial esophageal squamous cell carcinoma (ESCC) and precancerous lesions in endoscopic images based on the YOLOv5l model by using deep learning method of artificial intelligence to improve the diagnosis of early ESCC and precancerous lesions under endoscopy. 13, 009 endoscopic esophageal images of white light imaging (WLI), narrow band imaging (NBI) and lugol chromoendoscopy (LCE) were collected from June 2019 to July 2021 from 1, 126 patients at the Cancer Hospital, Chinese Academy of Medical Sciences, including low-grade intraepithelial neoplasia, high-grade intraepithelial neoplasia, ESCC limited to the mucosal layer, benign esophageal lesions and normal esophagus. By computerized random function method, the images were divided into a training set (11, 547 images from 1, 025 patients) and a validation set (1, 462 images from 101 patients). The YOLOv5l model was trained and constructed with the training set, and the model was validated with the validation set, while the validation set was diagnosed by two senior and two junior endoscopists, respectively, to compare the diagnostic results of YOLOv5l model and those of the endoscopists. In the validation set, the accuracy, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of the YOLOv5l model in diagnosing early ESCC and precancerous lesions in the WLI, NBI and LCE modes were 96.9%, 87.9%, 98.3%, 88.8%, 98.1%, and 98.6%, 89.3%, 99.5%, 94.4%, 98.2%, and 93.0%, 77.5%, 98.0%, 92.6%, 93.1%, respectively. The accuracy in the NBI model was higher than that in the WLI model (<0.05) and lower than that in the LCE model (<0.05). The diagnostic accuracies of YOLOv5l model in the WLI, NBI and LCE modes for the early ESCC and precancerous lesions were similar to those of the 2 senior endoscopists (96.9%, 98.8%, 94.3%, and 97.5%, 99.6%, 91.9%, respectively; >0.05), but significantly higher than those of the 2 junior endoscopists (84.7%, 92.9%, 81.6% and 88.3%, 91.9%, 81.2%, respectively; <0.05). The constructed YOLOv5l model has high accuracy in diagnosing early ESCC and precancerous lesions in endoscopic WLI, NBI and LCE modes, which can assist junior endoscopists to improve diagnosis and reduce missed diagnoses.
利用人工智能深度学习方法，基于YOLOv5l模型构建内镜图像中浅表性食管鳞状细胞癌（ESCC）及癌前病变的诊断模型，以提高内镜下早期ESCC及癌前病变的诊断水平。收集2019年6月至2021年7月中国医学科学院肿瘤医院1126例患者的13009幅白光成像（WLI）、窄带成像（NBI）及卢戈氏染色内镜检查（LCE）的食管内镜图像，包括低级别上皮内瘤变、高级别上皮内瘤变、局限于黏膜层的ESCC、食管良性病变及正常食管。通过计算机随机函数法将图像分为训练集（来自1025例患者的11547幅图像）和验证集（来自101例患者的1462幅图像）。用训练集对YOLOv5l模型进行训练和构建，并用验证集对模型进行验证，同时由两名高级和两名初级内镜医师分别对验证集进行诊断，以比较YOLOv5l模型与内镜医师的诊断结果。在验证集中，YOLOv5l模型在WLI、NBI和LCE模式下诊断早期ESCC及癌前病变的准确率、灵敏度、特异度、阳性预测值（PPV）和阴性预测值（NPV）分别为96.9%、87.9%、98.3%、88.8%、98.1%，98.6%、89.3%、99.5%、94.4%、98.2%，93.0%、77.5%、98.0%、92.6%、93.1%。NBI模式下的准确率高于WLI模式（<0.05），低于LCE模式（<0.05）。YOLOv5l模型在WLI、NBI和LCE模式下对早期ESCC及癌前病变的诊断准确率与两名高级内镜医师的诊断准确率相似（分别为96.9%、98.8%、94.3%和97.5%、99.6%、91.9%；>0.05），但显著高于两名初级内镜医师的诊断准确率（分别为84.7%、92.9%、81.6%和88.3%、91.9%、81.2%；<0.05）。构建的YOLOv5l模型在内镜WLI、NBI和LCE模式下诊断早期ESCC及癌前病变具有较高的准确率，可辅助初级内镜医师提高诊断水平并减少漏诊。
