Identification of upper GI diseases during screening gastroscopy using a deep convolutional neural network algorithm,Gastrointestinal Endoscopy

当前位置： X-MOL 学术 › Gastrointest. Endosc. › 论文详情

Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)

Identification of upper GI diseases during screening gastroscopy using a deep convolutional neural network algorithm
Gastrointestinal Endoscopy ( IF 7.7 ) Pub Date : 2022-06-16 , DOI: 10.1016/j.gie.2022.06.011
Hang Yang ₁ , Yu Wu ₂ , Bo Yang ₃ , Min Wu ₄ , Jun Zhou ₁ , Qin Liu ₁ , Yifei Lin ₅ , Shilin Li ₁ , Xue Li ₁ , Jie Zhang ₁ , Rui Wang ₁ , Qianrong Xie ₁ , Jingqi Li ₆ , Yue Luo ₇ , Mengjie Tu ₈ , Xiao Wang ₃ , Haitao Lan ₉ , Xuesong Bai ₃ , Huaping Wu ₁₀ , Fanwei Zeng ₁₁ , Hong Zhao ₁₂ , Zhang Yi ₂ , Fanxin Zeng ₁₃

Affiliation

Department of Clinical Research Center, Dazhou Central Hospital, Dazhou, Sichuan, China.
Center of Intelligent Medicine, Computer Science, Sichuan University, Chengdu, Sichuan, China.
Digestive Endoscopy Center, Dazhou Central Hospital, Dazhou, Sichuan, China.
Huaxi MR Research Center (HMRRC), Department of Radiology, West China Hospital, Sichuan University, Chengdu, Sichuan, China.
Precision Medicine Center, West China Hospital, Sichuan University, Chengdu, Sichuan, China.
College of Aulin, Northeast Forestry University, Harbin, Heilongjiang, China.
College of Basic Medical Sciences, North Sichuan Medical College, Nanchong, Sichuan, China.
Department of Clinical Research Center, Dazhou Central Hospital, Dazhou, Sichuan, China; Department of Surgery, Shantou University Medical College, Shantou, Guangdong, China.
Department of Sichuan, Academy of Medical Sciences, Sichuan Provincial People's Hospital, Chengdu, Sichuan, China.
Department of Cardiac &Vascular Surgery, Dazhou Central Hospital, Dazhou, Sichuan, China.
Department of Spinal Surgery, Sichuan Province Orthopedic Hospital, Chengdu, Sichuan, China.
Department of Hepatobiliary Surgery, State Key Laboratory of Molecular Oncology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences, Peking Union Medical College, Beijing, China.
Department of Clinical Research Center, Dazhou Central Hospital, Dazhou, Sichuan, China; Center of Intelligent Medicine, Computer Science, Sichuan University, Chengdu, Sichuan, China.

Background and Aims

The clinical application of GI endoscopy for the diagnosis of multiple diseases using artificial intelligence (AI) has been limited by its high false-positive rates. There is an unmet need to develop a GI endoscopy AI-assisted diagnosis system (GEADS) to improve diagnostic accuracy and clinical utility.

Methods

In this retrospective, multicenter study, a convolutional neural network was trained to assess upper GI diseases based on 26,228 endoscopic images from Dazhou Central Hospital that were randomly assigned (3:1:1) to a training dataset, validation dataset, and test dataset, respectively. To validate the model, 6 external independent datasets comprising 51,372 images of upper GI diseases were collected. In addition, 1 prospective dataset comprising 27,975 images was collected. The performance of GEADS was compared with endoscopists with 2 professional degrees of expertise: expert and novice. Eight endoscopists were in the expert group with >5 years of experience, whereas 3 endoscopists were in the novice group with 1 to 5 years of experience.

Results

The GEADS model achieved an accuracy of .918 (95% confidence interval [CI], .914-.922), with an F1 score of .884 (95% CI, .879-.889), recall of .873 (95% CI, .868-.878), and precision of .890 (95% CI, .885-.895) in the internal validation dataset. In the external validation datasets and 1 prospective validation dataset, the diagnostic accuracy of the GEADS ranged from .841 (95% CI, .834-.848) to .949 (95% CI, .935-.963). With the help of the GEADS, the diagnosing accuracies of novice and expert endoscopists were significantly improved (P < .001).

Conclusions

The AI system can assist endoscopists in improving the accuracy of diagnosing upper GI diseases.

中文翻译：

使用深度卷积神经网络算法在筛查胃镜检查中识别上消化道疾病

背景和目标

使用人工智能 (AI) 诊断多种疾病的胃肠道内窥镜的临床应用受到其高假阳性率的限制。开发 GI 内窥镜 AI 辅助诊断系统 (GEADS) 以提高诊断准确性和临床实用性的需求尚未得到满足。

方法

在这项回顾性、多中心研究中，基于来自达州市中心医院的 26,228 张内窥镜图像，卷积神经网络被训练以评估上消化道疾病，这些图像被随机分配 (3:1:1) 到训练数据集、验证数据集和测试数据集，分别。为了验证该模型，收集了 6 个外部独立数据集，包括 51,372 张上消化道疾病的图像。此外，还收集了 1 个包含 27,975 张图像的前瞻性数据集。GEADS 的表现与具有 2 个专业专业程度的内窥镜医师进行了比较：专家和新手。8 名内镜医师属于专家组，具有 >5 年的经验，而 3 名内镜医师属于新手组，具有 1 至 5 年的经验。

结果

GEADS 模型的准确度为 0.918（95% 置信区间 [CI]，0.914-.922），F1 得分为 0.884（95% CI，0.879-.889），召回率为 0.873（95 % CI，0.868-.878），内部验证数据集中的精度为 0.890（95% CI，0.885-.895）。在外部验证数据集和 1 个前瞻性验证数据集中，GEADS 的诊断准确性范围从 0.841（95% CI，0.834-.848）到 0.949（95% CI，0.935-.963）。在 GEADS 的帮助下，新手和专家内镜医师的诊断准确性显着提高 ( P < .001)。