Feng Jie, Zhang Yaoping, Feng Zhijun, Ma Huiming, Gou Yani, Wang Pengfei, Feng Yanhu, Wang Xiang, Huang Xiaojun
Department of Gastroenterology, Lanzhou University Second Hospital, Lanzhou, Gansu Province, China.
Digestive Endoscopy Engineering Research and Development Center of Gansu Province, Lanzhou, Gansu Province, China.
Surg Endosc. 2025 Mar;39(3):1874-1884. doi: 10.1007/s00464-025-11527-5. Epub 2025 Jan 22.
Endoscopic diagnosis of early gastric cancer (EGC) is a challenge. It is not clear whether deep convolutional neural network (DCNN) model could improve the endoscopists' diagnostic performance.
We established a DCNN-assisted system and found that accuracy of diagnosis is higher than endoscopists. We prospectively collected an independent image test set of 1289 images and a video test set of 130 patients from three endoscopic centers to compare the diagnostic efficacy of 12 endoscopists before and after DCNN model assistance. Accuracy, sensitivity, specificity, time, and AUC were the main indicators for comparison.
The DCNN model discriminated EGC from the control group (including ulcers and chronic gastritis) with an AUC of 0.917, a sensitivity of 93.38% (95% CI 91.09-95.12%), and a specificity of 90.07% (95% CI 87.59-92.10%) in the image dataset. The video test dataset have an AUC of 0.930, a sensitivity of 96.92% (95% CI 88.83-99.78%), and a specificity of 89.23% (95% CI 79.11-94.98%). The diagnostic performance of novice endoscopists was comparable to those of expert endoscopists with the DCNN model's assistance (accuracy: 95.22 vs. 96.16%) in image test dataset. In the video test, the novice endoscopists, accuracy after DCNN assistance was also improved from 79.36 to 86.41%, and from 86.28 to 91.03% for expert endoscopists. The mean pairwise kappa value of endoscopists was increased significantly with the DCNN model's assistance (0.705-0.753 vs.0.767-0.890) in image testing, and (0.657-0.793 vs. 0.738-0.905) in video testing. The diagnostic duration reduced considerably with the assistance of the DCNN model from 7.09 ± 0.6 s to 5.05 ± 0.55 s in image test, and from 2392.17 ± 7.77 s to2378.34 ± 23.51 s in video test.
We developed a DCNN-assisted diagnostic system. And the system can improve the diagnostic performance of endoscopists and help novice endoscopists achieve diagnostic accuracy comparable to that of expert endoscopists.
早期胃癌（EGC）的内镜诊断具有挑战性。目前尚不清楚深度卷积神经网络（DCNN）模型是否能提高内镜医师的诊断性能。
我们建立了一个DCNN辅助系统，发现其诊断准确率高于内镜医师。我们前瞻性地从三个内镜中心收集了一个包含1289张图像的独立图像测试集和一个包含130名患者的视频测试集，以比较DCNN模型辅助前后12名内镜医师的诊断效果。准确性、敏感性、特异性、诊断时间和曲线下面积（AUC）是主要的比较指标。
在图像数据集中，DCNN模型区分EGC与对照组（包括溃疡和慢性胃炎）的AUC为0.917，敏感性为93.38%（95%可信区间91.09 - 95.12%），特异性为90.07%（95%可信区间87.59 - 92.10%）。视频测试数据集的AUC为0.930，敏感性为96.92%（95%可信区间88.83 - 99.78%），特异性为89.23%（95%可信区间79.11 - 94.98%）。在DCNN模型的辅助下，新手内镜医师在图像测试数据集中的诊断性能与专家内镜医师相当（准确性：95.22%对96.16%）。在视频测试中，新手内镜医师在DCNN辅助后的准确性也从79.36%提高到86.41%，专家内镜医师从86.28%提高到91.03%。在图像测试中，内镜医师的平均配对kappa值在DCNN模型的辅助下显著增加（0.705 - 0.753对0.767 - 0.890），在视频测试中为（0.657 - 0.793对0.738 - 0.905）。在DCNN模型的辅助下，诊断时间大幅缩短，在图像测试中从7.09±0.6秒降至5.05±0.55秒，在视频测试中从2392.17±7.77秒降至2378.34±23.51秒。
我们开发了一种DCNN辅助诊断系统。该系统可以提高内镜医师的诊断性能，并帮助新手内镜医师达到与专家内镜医师相当的诊断准确性。
