使用人工智能通过磁控胶囊内镜检测胃部病变
本研究基于卷积神经网络(CNN)和快速区域卷积神经网络(RCNN)开发了自动胃部病变检测系统,使用797例患者的1,023,955张磁控胶囊内镜(MCE)图像进行训练和测试。系统检测胃局灶性病变的敏感度为96.2%(95%CI 95.7%-96.5%),特异度为76.2%(95%CI 75.97%-76.3%),准确率为77.1%(95%CI 76.9%-77.3%),受试者工作特征曲线下面积(AUC)为0.84。图像处理时间为每张44毫秒,显著快于临床医生的每张0.38±0.29秒(P<.001)。
要点
- 系统检测胃局灶性病变的敏感度为96.2%(95%CI 95.7%-96.5%),特异度为76.2%(95%CI 75.97%-76.3%)。
- 各病变类型敏感度:糜烂99.3%、息肉96.5%、溃疡89.3%、黏膜下肿瘤87.2%、黄色瘤90.6%、正常67.8%、无效图像96.1%。
- 系统处理每张图像44毫秒,临床医生每张0.38±0.29秒(P<.001)。
核心结论
- 基于CNN faster-RCNN的诊断程序系统在诊断MCE图像中的胃局灶性病变方面表现出良好性能。
研究设计与样本
- 研究基于卷积神经网络(CNN)和快速区域卷积神经网络(RCNN)开发了新型自动胃部病变检测系统。
- 共使用797例患者的1,023,955张MCE图像训练和测试系统,图像分为7类:糜烂、息肉、溃疡、黏膜下肿瘤、黄色瘤、正常黏膜和无效图像。
- 主要终点为系统的敏感度。
主要发现
- 系统检测胃局灶性病变的敏感度为96.2%(95%CI 95.7%-96.5%),特异度为76.2%(95%CI 75.97%-76.3%),阳性预测值为16.0%(95%CI 15.7%-16.3%),阴性预测值为99.7%(95%CI 99.74%-99.79%),准确率为77.1%(95%CI 76.9%-77.3%)。
- 各类型敏感度:糜烂99.3%、息肉96.5%、溃疡89.3%、黏膜下肿瘤87.2%、黄色瘤90.6%、正常67.8%、无效图像96.1%。
- 受试者工作特征曲线分析显示所有阳性图像的曲线下面积为0.84。
- 系统图像处理时间为每张44毫秒,临床医生为每张0.38±0.29秒(P<.001)。
- 重复读取2次的kappa值为1。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND AND AIMS: Magnetically controlled capsule endoscopy (MCE) has become an efficient diagnostic modality for gastric diseases. We developed a novel automatic gastric lesion detection system to assist in diagnosis and reduce inter-physician variations. This study aimed to evaluate the diagnostic capability of the computer-aided detection system for MCE images. METHODS: We developed a novel automatic gastric lesion detection system based on a convolutional neural network (CNN) and faster region-based convolutional neural network (RCNN). A total of 1,023,955 MCE images from 797 patients were used to train and test the system. These images were divided into 7 categories (erosion, polyp, ulcer, submucosal tumor, xanthoma, normal mucosa, and invalid images). The primary endpoint was the sensitivity of the system. RESULTS: The system detected gastric focal lesions with 96.2% sensitivity (95% confidence interval [CI], 95.7%-96.5%), 76.2% specificity (95% CI, 75.97%-76.3%), 16.0% positive predictive value (95% CI, 15.7%-16.3%), 99.7% negative predictive value (95% CI, 99.74%-99.79%), and 77.1% accuracy (95% CI, 76.9%-77.3%) (sensitivity was 99.3% for erosions; 96.5% for polyps; 89.3% for ulcers; 87.2% for submucosal tumors; 90.6% for xanthomas; 67.8% for normal; and 96.1% for invalid images). Analysis of the receiver operating characteristic curve showed that the area under the curve for all positive images was 0.84. Image processing time was 44 milliseconds per image for the system and 0.38 ± 0.29 seconds per image for clinicians (P < .001). The kappa value of 2 times repeated reads was 1. CONCLUSIONS: The CNN faster-RCNN-based diagnostic program system showed good performance in diagnosing gastric focal lesions in MCE images.
引用
- Xia J, Xia T, Pan J, et al. Use of artificial intelligence for detection of gastric lesions by magnetically controlled capsule endoscopy. Gastrointest Endosc. 2020. PMID: 32470426. doi:10.1016/j.gie.2020.05.027
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。