Real-time artificial intelligence for detection of upper gastrointestinal cancer by endoscopy: a multicentre, case-control, diagnostic study
Real-time artificial intelligence for detection of upper gastrointestinal cancer by endoscopy: a multicentre, case-control, diagnostic study
复制标题
DOI：
10.1016/s1470-2045(19)30637-0
复制
发表时间：
2019-12-01
期刊：
影响因子：
51.1
通讯作者：
Xu, Rui-hua
中科院分区：
文献类型：
--
作者：
Luo, Huiyan;Xu, Guoliang;Xu, Rui-hua
研究背景上消化道恶性肿瘤（包括食管癌和胃癌）是世界范围内最常见的恶性肿瘤。使用深度学习算法的人工智能平台在医学成像方面取得了显着进展，但它们在上消化道癌症中的应用有限。我们的目的是开发和验证胃肠道人工智能诊断系统（GRAIDS）的诊断上消化道癌症通过分析的影像数据，从临床endoscopy.Methods这多中心，病例对照，诊断研究在6所医院的不同级别（即，市，省，国家）在中国。从所有参与医院检索了年龄在18岁或以上、既往未接受过内窥镜检查的连续参与者的图像。所有经组织学证实为恶性肿瘤的上消化道癌病变（包括食管癌和胃癌）患者均符合本研究的条件。仅标准白色光图像被视为合格。将来自中山大学肿瘤防治中心的图像（8：1：1）随机分配到用于开发GRAIDS的训练和内部验证数据集以及用于评估GRAIDS性能的内部验证数据集。使用中山大学肿瘤中心（一家国立医院）的内部和前瞻性验证集以及来自五家初级保健医院的额外外部验证集评估其诊断性能。还将GRAIDS的性能与具有三种专业知识的内窥镜医生进行了比较：专家、合格和实习生。采用Clopper-Pearson方法计算95%CI，评价GRAIDS和内镜医师对癌性病变的诊断准确性、敏感性、特异性、阳性预测值和阴性预测值。在内部验证集中，识别上消化道癌症的诊断准确性为0.955（95%CI 0.952-0.957），在前瞻性集中为0.927（0.925-0.929），在5个外部验证集中的范围为0.915（0.913-0.917）至0.977（0.977-0.978）。GRAIDS的诊断灵敏度与专家内镜医师相似（0.942 [95% CI 0.924-0.957] vs 0.945 [0.927-0.959]; p=0.692），灵敏度上级合格内镜医师（0.858 [0.832-0.880]，p
Background Upper gastrointestinal cancers (including oesophageal cancer and gastric cancer) are the most common cancers worldwide. Artificial intelligence platforms using deep learning algorithms have made remarkable progress in medical imaging but their application in upper gastrointestinal cancers has been limited. We aimed to develop and validate the Gastrointestinal Artificial Intelligence Diagnostic System (GRAIDS) for the diagnosis of upper gastrointestinal cancers through analysis of imaging data from clinical endoscopies.Methods This multicentre, case-control, diagnostic study was done in six hospitals of different tiers (ie, municipal, provincial, and national) in China. The images of consecutive participants, aged 18 years or older, who had not had a previous endoscopy were retrieved from all participating hospitals. All patients with upper gastrointestinal cancer lesions (including oesophageal cancer and gastric cancer) that were histologically proven malignancies were eligible for this study. Only images with standard white light were deemed eligible. The images from Sun Yat-sen University Cancer Center were randomly assigned (8:1:1) to the training and intrinsic verification datasets for developing GRAIDS, and the internal validation dataset for evaluating the performance of GRAIDS. Its diagnostic performance was evaluated using an internal and prospective validation set from Sun Yat-sen University Cancer Center (a national hospital) and additional external validation sets from five primary care hospitals. The performance of GRAIDS was also compared with endoscopists with three degrees of expertise: expert, competent, and trainee. The diagnostic accuracy, sensitivity, specificity, positive predictive value, and negative predictive value of GRAIDS and endoscopists for the identification of cancerous lesions were evaluated by calculating the 95% CIs using the Clopper-Pearson method.Findings 1 036 496 endoscopy images from 84 424 individuals were used to develop and test GRAIDS. The diagnostic accuracy in identifying upper gastrointestinal cancers was 0.955 (95% CI 0.952-0.957) in the internal validation set, 0.927 (0.925-0.929) in the prospective set, and ranged from 0.915 (0.913-0.917) to 0.977 (0.977-0.978) in the five external validation sets. GRAIDS achieved diagnostic sensitivity similar to that of the expert endoscopist (0.942 [95% CI 0.924-0.957] vs 0.945 [0.927-0.959]; p=0.692) and superior sensitivity compared with competent (0.858 [0.832-0.880], p