浏览全部资源
扫码关注微信
基于深度学习的人工智能辅助胃镜下实时识别病变及位置模型的建立
Establishment of an artificial intelligence assisted diagnosis model based on deep learning for recognizing gastric lesions and their locations under gastroscopy in real time
- 局解手术学杂志 2024年33卷第10期 页码：849-853
- 作者机构：
陆军军医大学第二附属医院消化内科，重庆 400037
- 作者简介：
柏健鹰，E-mail：Drbaijy2015@163.com
- 基金信息：重庆市技术创新与应用发展专项重点项目(cstc2019jscx- gksbX0038)
DOI：10.11659/jjssx.02E024054
中图分类号： R445.9;TP183纸质出版日期：2024-10-25，
收稿日期：2024-02-26，
- 稿件说明：
移动端阅览
郭宪,吴应洋,江艾芮等.基于深度学习的人工智能辅助胃镜下实时识别病变及位置模型的建立[J].局解手术学杂志,2024,33(10):849-853.
GUO Xian,WU Ying-yang,JIANG Ai-rui,et al.Establishment of an artificial intelligence assisted diagnosis model based on deep learning for recognizing gastric lesions and their locations under gastroscopy in real time[J].JOURNAL OF REGIONAL ANATOMY AND OPERATIVE SURGERY,2024,33(10):849-853.
郭宪,吴应洋,江艾芮等.基于深度学习的人工智能辅助胃镜下实时识别病变及位置模型的建立[J].局解手术学杂志,2024,33(10):849-853. DOI： 10.11659/jjssx.02E024054.
GUO Xian,WU Ying-yang,JIANG Ai-rui,et al.Establishment of an artificial intelligence assisted diagnosis model based on deep learning for recognizing gastric lesions and their locations under gastroscopy in real time[J].JOURNAL OF REGIONAL ANATOMY AND OPERATIVE SURGERY,2024,33(10):849-853. DOI： 10.11659/jjssx.02E024054.
摘要
目的
2
构建一个基于深度学习的人工智能辅助诊断模型，用于实时动态识别胃镜下的胃部病变及位置，并评估其对胃部病变检出和位置识别的能力。
方法
2
回顾性分析我院104例患者的胃镜检查视频，对视频帧进行人工标注，将已标注的病变类别图片帧按8：2的比例划分成训练集与验证集，将已标注的位置类别图片帧根据患者来源按8：2的比例划分成训练集与验证集，分别用于模型的训练及验证。病变识别部分的训练采用YoloV4模型，位置识别部分的训练采用ResNet152模型。评估辅助诊断模型对病变识别的准确率、敏感度、特异度、阳性预测值、阴性预测值，及对位置识别的准确率。
结果
2
共标注图片68 351帧，其中训练集图片54 872帧，包括病变类别41 692帧，位置类别13 180帧；验证集图片13 479帧，包括病变类别10 422帧，位置类别3 057帧。在验证集中，病变识别模型的总体准确率为98.8%，敏感度为96.6%，特异度为99.3%，阳性预测值为96.3%，阴性预测值为99.3%；位置识别模型top5总体准确率为87.1%。
结论
2
基于深度学习的人工智能辅助诊断模型用于实时动态识别胃镜下胃部病变及位置有较好的病变检出能力和位置识别能力，临床应用前景巨大。
Abstract
Objective
2
To construct an artificial intelligence assisted diagnosis model based on deep learning for dynamically recognizing gastric lesions and their locations under gastroscopy in real time， and to evaluate its ability to detect and recognize gastric lesions and their locations.
Methods
2
The gastroscopy videos of 104 patients in our hospital was retrospectively analyzed， and the video frames were manually annotated. The annotated picture frames of lesion category were divided into the training set and the validation set according to the ratio of 8∶2， and the annotated picture frames of location category were divided into the training set and the validation set according to the patient sources at the ratio of 8∶2.These sets were utilized for training and validating the respective models.YoloV4 model was used for the training of lesion recognition， and ResNet152 model was used for the training of location recognition.The accuracy， sensitivity， specificity， positive predictive value， negative predictive value and location recognition accuracy of the auxiliary diagnostic model were evaluated.
Results
2
A total of 68 351 image frames were annotated，with 54 872 frames used as the training set，including 41 692 frames for lesion categories and 13 180 frames for location categories.The validation set consisted of 13 479 frames，comprising 10 422 frames for lesion categories and 3 057 frames for location categories. The lesion recognition model achieved an overall accuracy of 98.8%，with a sensitivity of 96.6%，specificity of 99.3%，positive predictive value of 96.3%，and negative predictive value of 99.3% in validation set.Meanwhile，the location recognition model demonstrated an top-5 accuracy of 87.1%.
Conclusion
2
The artificial intelligence assisted diagnosis model based on deep learning for real-time dynamic recognition of gastric lesions and their locations under gastroscopy has good ability in lesion detection and location recognition， and has great clinical application prospects.
关键词
人工智能胃部疾病早期胃癌深度学习胃镜检查
Keywords
artificial intelligencegastric diseasesearly gastric cancerdeep learninggastroscopy examination
references
何敏均，季璐玓，连力，等.2000-2019年中国早发型胃癌和晚发型胃癌流行病学趋势分析［J］.中华流行病学杂志，2023，44（8）：1198-1202.doi：10.3760/cma.j.cn112338-20230302-00117http://dx.doi.org/10.3760/cma.j.cn112338-20230302-00117.
郑荣寿，陈茹，韩冰峰，等.2022年中国恶性肿瘤流行情况分析［J］.中华肿瘤杂志，2024，46（3）：221-231.doi：10.3760/cma.j.cn112152-20240119-00035http://dx.doi.org/10.3760/cma.j.cn112152-20240119-00035.
Pimenta-Melo AR，Monteiro-Soares M，Libânio D，et al.Missing rate for gastric cancer during upper gastrointestinal endoscopy：a systematic review and meta-analysis［J］.Eur J Gastroenterol Hepatol，2016，28（9）：1041-1049.doi：10.1097/MEG.0000000000000657http://dx.doi.org/10.1097/MEG.0000000000000657.
张梦娇，徐铭，吴练练，等.内镜人工智能诊断辅助系统对胃局灶性病变检出的应用（含视频）［J］.中华消化内镜杂志，2023，40（5）：372-378.doi：10.3760/cma.j.cn321463-20221022-00575http://dx.doi.org/10.3760/cma.j.cn321463-20221022-00575.
王洛伟，马旭东，李兆申.加强消化内镜诊疗质控 推动消化内镜高质量发展［J］.中华消化内镜杂志，2022，39（11）：857-859.doi：10.3760/cma.j.cn321463-20220411-00043http://dx.doi.org/10.3760/cma.j.cn321463-20220411-00043.
高野，冯拥璞，刘雨，等.新时代消化内镜技术培训和内镜医师培养［J］.中华消化内镜杂志，2020，37（1）：3-10.doi：10.3760/cma.j.issn.1007-5232.2020.01.002http://dx.doi.org/10.3760/cma.j.issn.1007-5232.2020.01.002.
徐铭，姚理文，于红刚.智能消化内镜质控系统在胃镜检查中的应用价值研究［J］.中华消化内镜杂志，2022，39（2）：133-138.doi：10.3760/cma.j.cn321463-20200426-00358http://dx.doi.org/10.3760/cma.j.cn321463-20200426-00358.
LeCun Y，Bengio Y，Hinton G.Deep learning［J］.Nature，2015，521（7553）：436-444.doi：10.1038/nature14539http://dx.doi.org/10.1038/nature14539.
王瑜，吴江，伍东升.卷积神经网络在职业性尘肺病影像学诊断中的应用研究进展［J］.生物医学工程学杂志，2024，41（2）：413-420.doi：10.7507/1001-5515.202309079http://dx.doi.org/10.7507/1001-5515.202309079.
Khozin S.From organs to algorithms：redefining cancer classification in the age of artificial intelligence［J］.Clin Transl Sci，2024，17（9）：e70001.doi：10.1111/cts.70001http://dx.doi.org/10.1111/cts.70001.
Tognetti L，Miracapillo C，Leonardelli S，et al.Deep learning techniques for the dermoscopic differential diagnosis of benign/malignant melanocytic skin lesions：from the past to the present［J］.Bioengineering （Basel），2024，11（8）：758.doi：10.3390/bioengineering 11080758http://dx.doi.org/10.3390/bioengineering11080758.
Ruamviboonsuk P，Cheung CY，Zhang X，et al.Artificial intelligence in ophthalmology：evolutions in Asia［J］.Asia Pac J Ophthalmol （Phila），2020，9（2）：78-84.doi：10.1097/01.APO.0000656980. 41190.bfhttp://dx.doi.org/10.1097/01.APO.0000656980.41190.bf.
Ribeiro AH，Ribeiro MH，Paixão GMM，et al.Automatic diagnosis of the 12-lead ECG using a deep neural network［J］.Nat Commun，2020，11（1）：1760.doi：10.1038/s41467-020-15432-4http://dx.doi.org/10.1038/s41467-020-15432-4.
Bellemo V，Lim G，Rim TH，et al.Artificial intelligence screening for diabetic retinopathy：the real-world emerging application［J］.Curr Diab Rep，2019，19（9）：72.doi：10.1007/s11892-019-1189-3http://dx.doi.org/10.1007/s11892-019-1189-3.
Miyaki R，Yoshida S，Tanaka S，et al.A computer system to be used with laser-based endoscopy for quantitative diagnosis of early gastric cancer［J］.J Clin Gastroenterol，2015，49（2）：108-115.doi：10.1097/MCG.0000000000000104http://dx.doi.org/10.1097/MCG.0000000000000104.
Wu L，Zhou W，Wan X，et al.A deep neural network improves endoscopic detection of early gastric cancer without blind spots［J］.Endoscopy，2019，51（6）：522-531.doi：10.1055/a-0855-3532http://dx.doi.org/10.1055/a-0855-3532.
黄丽，李艳霞，吴练练，等.基于深度学习的良恶性胃溃疡人工智能辅助诊断系统研究［J］.中华消化内镜杂志，2020，37（7）：476-480.doi：10.3760/cma.j.cn321463-20200105-00547http://dx.doi.org/10.3760/cma.j.cn321463-20200105-00547.
Hamashima C，Shibuya D，Yamazaki H，et al.The Japanese guidelines for gastric cancer screening［J］.Jpn J Clin Oncol，2008，38（4）：259-267.doi：10.1093/jjco/hyn017http://dx.doi.org/10.1093/jjco/hyn017.
Gao QY，Fang JY.Early esophageal cancer screening in China［J］.Best Pract Res Clin Gastroenterol，2015，29（6）：885-893.doi：10.1016/j.bpg.2015.09.018http://dx.doi.org/10.1016/j.bpg.2015.09.018.
Xie Y，Yu Y，Liao M，et al.Gastric polyp detection module based on improved attentional feature fusion［J］.Biomed Eng Online，2023，22（1）：72.doi：10.1186/s12938-023-01130-xhttp://dx.doi.org/10.1186/s12938-023-01130-x.
赵曲川，池添雨.基于U-Net深度学习慢性萎缩性胃炎模型的应用与研究［J］.胃肠病学和肝病学杂志，2022，31（6）：656-661.doi：10.3969/j.issn.1006-5709.2022.06.010http://dx.doi.org/10.3969/j.issn.1006-5709.2022.06.010.
Hu H，Gong L，Dong D，et al.Identifying early gastric cancer under magnifying narrow-band images with deep learning：a multicenter study［J］.Gastrointest Endosc，2021，93（6）：1333-1341.e3.doi：10.1016/j.gie.2020.11.014http://dx.doi.org/10.1016/j.gie.2020.11.014.
Sun Y，Yin L，Nesheli DN，et al.Overall and cause-specific mortality among patients diagnosed with gastric precancerous lesions in Sweden between 1979 and 2014：an observational cohort study［J］.BMC Med，2024，22（1）：333.doi：10.1186/s12916-024-03554-1http://dx.doi.org/10.1186/s12916-024-03554-1.
张希钢，赖春晓，戴捷，等.Transformer模型在胃镜视频部位实时识别的研究探索［J］.现代消化及介入诊疗，2022，27（1）：7-11.doi：10.3969/j.issn.1672-2159.2022.01.003http://dx.doi.org/10.3969/j.issn.1672-2159.2022.01.003.
0
浏览量
0
下载量
0
CSCD
- 网络PDF下载
- WORD下载
- Meta-XML下载
- BibTeX下载
- Endnote下载
- RefMan 下载
- NoteExpress下载
- NoteFirst下载
关联资源
相关文章
相关作者
相关机构
- 技术支持由北京北大方正电子有限公司提供 京ICP备09064830号-19
京公网安备11010802024621 - 本系统建议在Chrome、 IE9+ 以上版本浏览器阅读本站内容，360浏览器请切换至极速模式
- Cookies帮助我们提供服务并提供个性化体验。使用本网站，即表示您同意我们使用Cookies
