Tang Dehua, Zhou Jie, Wang Lei, Ni Muhan, Chen Min, Hassan Shahzeb, Luo Renquan, Chen Xi, He Xinqi, Zhang Lihui, Ding Xiwei, Yu Honggang, Xu Guifang, Zou Xiaoping
Department of Gastroenterology, Nanjing Drum Tower Hospital, Affiliated Drum Tower Hospital, Medical School of Nanjing University, Nanjing, China.
Department of Gastroenterology, Renmin Hospital of Wuhan University, Wuhan, China.
Front Oncol. 2021 Apr 20;11:622827. doi: 10.3389/fonc.2021.622827. eCollection 2021.
Prediction of intramucosal gastric cancer (GC) is a big challenge. It is not clear whether artificial intelligence could assist endoscopists in the diagnosis.
A deep convolutional neural networks (DCNN) model was developed retrospectively collected 3407 endoscopic images from 666 gastric cancer patients from two Endoscopy Centers (training dataset). The DCNN model's performance was tested with 228 images from 62 independent patients (testing dataset). The endoscopists evaluated the image and video testing dataset with or without the DCNN model's assistance, respectively. Endoscopists' diagnostic performance was compared with or without the DCNN model's assistance and investigated the effects of assistance using correlations and linear regression analyses.
The DCNN model discriminated intramucosal GC from advanced GC with an AUC of 0.942 (95% CI, 0.915-0.970), a sensitivity of 90.5% (95% CI, 84.1%-95.4%), and a specificity of 85.3% (95% CI, 77.1%-90.9%) in the testing dataset. The diagnostic performance of novice endoscopists was comparable to those of expert endoscopists with the DCNN model's assistance (accuracy: 84.6% 85.5%, sensitivity: 85.7% 87.4%, specificity: 83.3% 83.0%). The mean pairwise kappa value of endoscopists was increased significantly with the DCNN model's assistance (0.430-0.629 0.660-0.861). The diagnostic duration reduced considerably with the assistance of the DCNN model from 4.35s to 3.01s. The correlation between the perseverance of effort and diagnostic accuracy of endoscopists was diminished using the DCNN model (r: 0.470 0.076).
An AI-assisted system was established and found useful for novice endoscopists to achieve comparable diagnostic performance with experts.
黏膜内胃癌（GC）的预测是一项巨大挑战。目前尚不清楚人工智能是否能协助内镜医师进行诊断。
开发了一种深度卷积神经网络（DCNN）模型，回顾性收集了来自两个内镜中心的666例胃癌患者的3407张内镜图像（训练数据集）。使用来自62例独立患者的228张图像（测试数据集）对DCNN模型的性能进行测试。内镜医师分别在有或没有DCNN模型协助的情况下评估图像和视频测试数据集。比较了有或没有DCNN模型协助时内镜医师的诊断性能，并使用相关性和线性回归分析研究了协助的效果。
在测试数据集中，DCNN模型区分黏膜内GC和进展期GC的AUC为0.942（95%CI，0.915 - 0.970），灵敏度为90.5%（95%CI，84.1% - 95.4%），特异度为85.3%（95%CI，77.1% - 90.9%）。在DCNN模型的协助下，新手内镜医师的诊断性能与专家内镜医师相当（准确率：84.6%对85.5%，灵敏度：85.7%对87.4%，特异度：83.3%对83.0%）。在DCNN模型的协助下，内镜医师的平均两两kappa值显著增加（0.430 - 0.629对0.660 - 0.861）。在DCNN模型的协助下，诊断时间从4.35秒大幅缩短至3.01秒。使用DCNN模型后，内镜医师的努力坚持与诊断准确性之间的相关性降低（r：0.470对0.076）。
建立了一种人工智能辅助系统，发现其对新手内镜医师有用，可使其获得与专家相当的诊断性能。
