基于卷积神经网络的人工智能技术在早期胃癌识别中的应用

吴宏博; 姚幸雨; 曾丽莎; 黄访; 陈磊

doi:10.16016/j.1000-5404.202105018

您当前的位置：

首页 >

文章列表页 >

基于卷积神经网络的人工智能技术在早期胃癌识别中的应用

专题报道：数字医学与人工智能 | 更新时间：2026-01-09

- 基于卷积神经网络的人工智能技术在早期胃癌识别中的应用
- Application of artificial intelligence technology based on convolutional neural network in early gastric cancer recognition
- 第三军医大学学报 2021年43卷第18期页码：1735-1742
- 作者机构：
  
  400038 重庆，陆军军医大学第一附属医院消化科
- 作者简介：
  
  [陈磊, E-mail: chenlei1977603@126.comCHEN Lei, E-mail: chenlei1977603@126.com
- 基金信息：
- DOI：10.16016/j.1000-5404.202105018
  中图分类号： R319;R730.49;R735.2
- 收稿：2021-05-07，
  
  修回：2021-07-26，
  
  纸质出版：2021-09-30
- 稿件说明：
移动端阅览
吴宏博, 姚幸雨, 曾丽莎, 等. 基于卷积神经网络的人工智能技术在早期胃癌识别中的应用[J]. 第三军医大学学报, 2021,43(18):1735-1742.

Hongbo WU, Xingyu YAO, Lisha ZENG, et al. Application of artificial intelligence technology based on convolutional neural network in early gastric cancer recognition[J]. Journal of Third Military Medical University, 2021, 43(18): 1735-1742.
吴宏博, 姚幸雨, 曾丽莎, 等. 基于卷积神经网络的人工智能技术在早期胃癌识别中的应用[J]. 第三军医大学学报, 2021,43(18):1735-1742. DOI： 10.16016/j.1000-5404.202105018.

Hongbo WU, Xingyu YAO, Lisha ZENG, et al. Application of artificial intelligence technology based on convolutional neural network in early gastric cancer recognition[J]. Journal of Third Military Medical University, 2021, 43(18): 1735-1742. DOI： 10.16016/j.1000-5404.202105018.

摘要

目的

构建及验证一个用于早期胃癌识别的卷积神经网络模型，旨在提高早期胃癌的检出率。

方法

从陆军军医大学西南医院内镜中心数据库收集2016年1月至2020年8月的电子胃镜图片和胃镜检查视频。选取928例患者共5 496张包含早期胃癌、良性病变和正常的图片。随机分为训练集(662例患者共4 167张包含早期胃癌、良性病变和正常的图片)，验证集(259例患者共1 329张包含早期胃癌、良性病变和正常的图片)用于模型和4名内镜医师进行识别，最后统计相关结果。再从模型识别出的胃早癌图片中随机选取458张用作模型定位准确度的分析，分析模型框出病灶面积与内镜专家标注面积的重合度。将34例胃部内镜检查视频用于模型进行实时识别，模型识别出的视频病灶图片经内镜专家审核后，得出模型识别早期胃癌的灵敏性指标。

结果

模型用于早期胃癌的识别灵敏度和阳性预测值(PPV)分别是90.33%，95.41%，识别慢性浅表性胃炎、胃息肉和胃溃疡等良性病变的灵敏度和PPV均超过80%，单张识别时间为(0.04±0.005)s。在对早癌的检出率和诊断时间方面，模型均优于内镜医师组；经卡方检验，在早癌病灶的识别上优于内镜医师组。在模型准确定位方面，模型识别重合度达到60%以上的图片张数有380张，占总数82.97%；就形态而言，模型对隆起型病灶定位最准确，不同规格重合度(重合度≧60%、重合度≧70%)的图片数都多于平坦和凹陷型病灶。在视频验证实验中，模型从19处胃早癌病灶中，正确识别出17处，灵敏度为89.5%，对早癌病灶的检出率与活检病理证实的早癌检出率吻合度较强，具有良好的一致性。

结论

本文搭建的卷积神经网络模型在识别早期胃癌和三种良性病变(慢性浅表性胃炎、胃息肉和胃溃疡)具有较高的灵敏度和阳性预测值，能精确定位早癌病灶位置和边界，同时还能对早期胃癌及良性病变进行动态识别，能在实际临床检查中辅助内镜医师提高早期胃癌检出率，提升诊断水平。

Abstract

Objective

To construct and verify a convolutional neural network model for early gastric cancer recognition in order to improve the detection rate of early gastric cancer.

Methods

We collected the gastroscopy stock photos and gastroscopy videos in the database of the Endoscopy Center of our hospital from January 2016 to August 2020. A total of 5 496 photos from 928 patients were subjected

including early gastric cancer

benign lesions and normal pictures. The photos were randomly divided into training set (662 patients

4 167 photos)

and validation set (259 patients

1 329 photos). The model was identified with 4 endoscopists

and finally the relevant Results were counted. Then

458 photos of early gastric cancer identified by the model were randomly selected for the analysis of the accuracy of model positioning

and the overlap between the area of the lesion framed by the model and the area marked by the endoscopy experts was analyzed. Thirty-four cases of gastric endoscopy videos were used for real-time identification of the model. After the video lesion photos identified by the model were reviewed by endoscopic experts

the sensitivity index for the model to identify early gastric cancer was obtained.

Results

The recognition sensitivity and positive predictive value (PPV) of the model for early gastric cancer were 90.33% and 95.41%

respectively. The sensitivity and PPV for identifying benign lesions such as chronic superficial gastritis

gastric polyps and gastric ulcers were both over 80%. The sheet recognition time was 0.040±0.005 s. In terms of the detection rate and diagnosis time of early cancer

the model was better than the endoscopist group; after

Chi

-square test

it was better than the endoscopist group in the identification of early cancer lesions. In terms of accurate positioning of the model

there were 380 photos with an overlap of more t

han 60% of the model recognition

accounting for 82.97% of the total. In terms of morphology

the model showed most accurate for positioning uplifted lesions

and the number of picture of the overlap of different specifications (coincidence ≥60 %

coincidence ≥70%) were larger than flat and depressed lesions. In the video verification experiment

the model correctly identified 17 of the 19 early gastric cancer lesions with a sensitivity of 89.5%. The detection rate of early cancer lesions was in good consistency with the early cancer detection rate confirmed by biopsy pathology.

Conclusion

Our constructed convolutional neural network model presents high sensitivity and positive predictive value in the identification of early gastric cancer and 3 benign lesions (chronic superficial gastritis

gastric polyps

and gastric ulcer)

and can accurately identify the location and margin of early cancer lesions. At the same time

it can also dynamically identify early gastric cancer and benign lesions

which can assist endoscopists in the actual clinical examination to increase the detection rate of early gastric cancer and improve the level of diagnosis.

关键词

Keywords

references

CHEN W J, ZHENG R S, Baade P D, et al . Cancer statistics in China, 2015[J ] . CA Cancer J Clin, 2016, 66(2): 115-132.

HIRASAWA T, AOYAMA K, TANIMOTO T, et al . Application of artificial intelligence using a convolutional neural network for detecting gastric cancer in endoscopic images[J ] . Gastric Cancer, 2018, 21(4): 653-660. DOI: 10.1007/s10120-018-0793-2.

蔡世伦, 阿依木克地斯·亚力孔, 李染等. 基于深度学习的人工智能辅助诊断在食管早癌中的应用[J]. 中华消化内镜杂志, 2019, 36(4): 246-250.

CAI S L, AYIMUKEDISI Y, LI R, et al . Application of artificial intelligence assisted diagnosis based on deep learning for early esophageal cancer[J ] . Chin J Digestive Endosc, 2019, 36(4): 246-250.

BANG C S, BAIK G H, KIM J H, et al . Effect of training in upper endoscopic biopsy[J ] . Korean J Helicobacter Up Gastrointest Res, 2015, 15(1): 33. DOI: 10.7704/kjhugr.2015.15.1.33.

MUGURUMA N, MIYAMOTO H, OKAHIS A T, et al . Endoscopic molecular imaging: status and future perspective[J ] . Clin Endosc, 2013, 46(6): 603-610. DOI: 10.5946/ce.2013.46.6.603.

COHEN J, SAFDI M A, DEAL S E, et al . Quality indicators for esophagogastroduodenoscopy[J ] . Am J Gastroenterol, 2006, 101(4): 886-891. DOI: 10.1111/j.1572-0241.2006.00676.x.

IKENOYAMA Y, HIRASAWA T, ISHIOKA M, et al . Detecting early gastric cancer: Comparison between the diagnostic ability of convolutional neural networks andendoscopists[J ] . Dig Endosc, 2021, 33(1): 141-150. DOI: 10.1111/den.13688.

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.

王智杰, 高杰, 孟茜茜, 等. 基于深度学习的人工智能技术在早期胃癌诊断中的应用[J]. 中华消化内镜杂志, 2018, 35(8): 551-556. DOI: 10.3760/cma.j.issn.1007-5232.2018.08.004.

WANG Z J, GAO J, MENG Q Q, et al . Artificial intelligence based on deep learning for automatic detection of early gastric cancer[J ] . Chin J Digestive Endosc, 2018, 35(8): 551-556. DOI: 10.3760/cma.j.issn.1007-5232.2018.08.004.

LI Y P, DENG L Y, YANG X H, et al . Early diagnosis of gastric cancer based on deep learning combined with the spectral-spatial classification method[J ] . Biomed Opt Express, 2019, 10(10): 4999-5014. DOI: 10.1364/BOE.10.004999.

GUIMARÃES P, KELLER A, FEHLMANN T, et al . Deep-learning based detection of gastric precancerous conditions[J ] . Gut, 2020, 69(1): 4-6. DOI: 10.1136/gutjnl-2019-319347.

ZHU Y, WANG Q C, XU M D, et al . Application of convolutional neural network in the diagnosis of the invasion depth of gastric cancer based on conventional endoscopy[J ] . Gastrointest Endosc, 2019, 89(4): 806-815.e1. DOI: 10.1016/j.gie.2018.11.011.

ZHANG X, HU W L, CHEN F, et al . Gastric precancerous diseases classification using CNN with a concise model[J ] . PLoS ONE, 2017, 12(9): e0185508. DOI: 10.1371/journal.pone.0185508.

LI L, CHEN Y S, SHEN Z, et al . Convolutional neural network for the diagnosis of early gastric cancer based on magnifying narrow band imaging[J ] . Gastric Cancer, 2020, 23(1): 126-132. DOI: 10.1007/s10120-019-00992-2.

HORIUCHI Y, AOYAMA K, TOKAI Y, et al . Convolutional neural network for differentiating gastric cancer from gastritis using magnified endoscopy with narrow band imaging[J ] . Dig Dis Sci, 2020, 65(5): 1355-1363. DOI: 10.1007/s10620-019-05862-6.

KANESAKA T, LEE T C, UEDO N, et al . Computer-aided diagnosis for identifying and delineating early gastric cancers in magnifying narrow-band imaging[J ] . Gastrointest Endosc, 2018, 87(5): 1339-1344. DOI: 10.1016/j.gie.2017.11.029.

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.0000000000000104.

CHO B J, BANG C S, PARK S W, et al . Automated classification of gastric neoplasms in endoscopic images using a convolutional neural network[J ] . Endoscopy, 2019, 51(12): 1121-1129. DOI: 10.1055/a-0981-6133.

LEE J H, KIM Y J, KIM Y W, et al . Spotting malignancies from gastric endoscopic images using deep learning[J ] . Surg Endosc, 2019, 33(11): 3790-3797. DOI: 10.1007/s00464-019-06677-2.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

奥沙利铂通过诱导胃癌细胞铜死亡抑制胃癌

LncRNA FGD5-AS1通过调节miR-302b-3p/E2F1轴对胃癌细胞迁移及侵袭的影响

转录因子TGIF1高表达胃癌的免疫抑制性微环境特征及临床意义分析

内镜下LCPL征象对胃黏膜高危肠上皮化生的诊断价值

钙腔蛋白对胃癌迁移、侵袭及患者预后的影响