{"title":"基于深度学习的虚拟色内镜在胃肿瘤中的诊断性能。","authors":"Sho Suzuki, Yusuke Monno, Ryo Arai, Masaki Miyaoka, Yosuke Toya, Mitsuru Esaki, Takuya Wada, Waku Hatta, Ayaka Takasu, Shigeaki Nagao, Fumiaki Ishibashi, Yohei Minato, Kenichi Konda, Takahiro Dohmen, Kenji Miki, Masatoshi Okutomi","doi":"10.1007/s10120-024-01469-7","DOIUrl":null,"url":null,"abstract":"<p><strong>Backgrounds: </strong>Cycle-consistent generative adversarial network (CycleGAN) is a deep neural network model that performs image-to-image translations. We generated virtual indigo carmine (IC) chromoendoscopy images of gastric neoplasms using CycleGAN and compared their diagnostic performance with that of white light endoscopy (WLE).</p><p><strong>Methods: </strong>WLE and IC images of 176 patients with gastric neoplasms who underwent endoscopic resection were obtained. We used 1,633 images (911 WLE and 722 IC) of 146 cases in the training dataset to develop virtual IC images using CycleGAN. The remaining 30 WLE images were translated into 30 virtual IC images using the trained CycleGAN and used for validation. The lesion borders were evaluated by 118 endoscopists from 22 institutions using the 60 paired virtual IC and WLE images. The lesion area concordance rate and successful whole-lesion diagnosis were compared.</p><p><strong>Results: </strong>The lesion area concordance rate based on the pathological diagnosis in virtual IC was lower than in WLE (44.1% vs. 48.5%, p < 0.01). The successful whole-lesion diagnosis was higher in the virtual IC than in WLE images; however, the difference was insignificant (28.2% vs. 26.4%, p = 0.11). Conversely, subgroup analyses revealed a significantly higher diagnosis in virtual IC than in WLE for depressed morphology (41.9% vs. 36.9%, p = 0.02), differentiated histology (27.6% vs. 24.8%, p = 0.02), smaller lesion size (42.3% vs. 38.3%, p = 0.01), and assessed by expert endoscopists (27.3% vs. 23.6%, p = 0.03).</p><p><strong>Conclusions: </strong>The diagnostic ability of virtual IC was higher for some lesions, but not completely superior to that of WLE. Adjustments are required to improve the imaging system's performance.</p>","PeriodicalId":12684,"journal":{"name":"Gastric Cancer","volume":null,"pages":null},"PeriodicalIF":6.0000,"publicationDate":"2024-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Diagnostic performance of deep-learning-based virtual chromoendoscopy in gastric neoplasms.\",\"authors\":\"Sho Suzuki, Yusuke Monno, Ryo Arai, Masaki Miyaoka, Yosuke Toya, Mitsuru Esaki, Takuya Wada, Waku Hatta, Ayaka Takasu, Shigeaki Nagao, Fumiaki Ishibashi, Yohei Minato, Kenichi Konda, Takahiro Dohmen, Kenji Miki, Masatoshi Okutomi\",\"doi\":\"10.1007/s10120-024-01469-7\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p><strong>Backgrounds: </strong>Cycle-consistent generative adversarial network (CycleGAN) is a deep neural network model that performs image-to-image translations. We generated virtual indigo carmine (IC) chromoendoscopy images of gastric neoplasms using CycleGAN and compared their diagnostic performance with that of white light endoscopy (WLE).</p><p><strong>Methods: </strong>WLE and IC images of 176 patients with gastric neoplasms who underwent endoscopic resection were obtained. We used 1,633 images (911 WLE and 722 IC) of 146 cases in the training dataset to develop virtual IC images using CycleGAN. The remaining 30 WLE images were translated into 30 virtual IC images using the trained CycleGAN and used for validation. The lesion borders were evaluated by 118 endoscopists from 22 institutions using the 60 paired virtual IC and WLE images. The lesion area concordance rate and successful whole-lesion diagnosis were compared.</p><p><strong>Results: </strong>The lesion area concordance rate based on the pathological diagnosis in virtual IC was lower than in WLE (44.1% vs. 48.5%, p < 0.01). The successful whole-lesion diagnosis was higher in the virtual IC than in WLE images; however, the difference was insignificant (28.2% vs. 26.4%, p = 0.11). Conversely, subgroup analyses revealed a significantly higher diagnosis in virtual IC than in WLE for depressed morphology (41.9% vs. 36.9%, p = 0.02), differentiated histology (27.6% vs. 24.8%, p = 0.02), smaller lesion size (42.3% vs. 38.3%, p = 0.01), and assessed by expert endoscopists (27.3% vs. 23.6%, p = 0.03).</p><p><strong>Conclusions: </strong>The diagnostic ability of virtual IC was higher for some lesions, but not completely superior to that of WLE. Adjustments are required to improve the imaging system's performance.</p>\",\"PeriodicalId\":12684,\"journal\":{\"name\":\"Gastric Cancer\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":6.0000,\"publicationDate\":\"2024-05-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Gastric Cancer\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://doi.org/10.1007/s10120-024-01469-7\",\"RegionNum\":1,\"RegionCategory\":\"医学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"2024/1/19 0:00:00\",\"PubModel\":\"Epub\",\"JCR\":\"Q1\",\"JCRName\":\"GASTROENTEROLOGY & HEPATOLOGY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Gastric Cancer","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1007/s10120-024-01469-7","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2024/1/19 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"GASTROENTEROLOGY & HEPATOLOGY","Score":null,"Total":0}
引用次数: 0
摘要
背景:循环一致性生成对抗网络(CycleGAN)是一种深度神经网络模型,可进行图像到图像的转换。我们利用 CycleGAN 生成了虚拟的靛蓝胭脂红(IC)胃肿瘤色内镜图像,并将其诊断性能与白光内镜(WLE)进行了比较:方法: 我们获得了 176 名接受内镜切除术的胃肿瘤患者的 WLE 和 IC 图像。我们使用训练数据集中 146 个病例的 1,633 张图像(911 张 WLE 和 722 张 IC),利用 CycleGAN 开发了虚拟 IC 图像。剩下的 30 张 WLE 图像则使用训练好的 CycleGAN 转换成 30 张虚拟 IC 图像,并用于验证。来自 22 个机构的 118 名内镜医师使用 60 张配对的虚拟 IC 和 WLE 图像对病灶边界进行了评估。比较了病变区域吻合率和全病变诊断成功率:结果:根据病理诊断,虚拟 IC 的病灶面积吻合率低于 WLE(44.1% 对 48.5%,P 结论:虚拟 IC 的病灶面积吻合率低于 WLE(44.1% 对 48.5%,P 结论):虚拟 IC 对某些病变的诊断能力较高,但并不完全优于 WLE。需要进行调整以提高成像系统的性能。
Diagnostic performance of deep-learning-based virtual chromoendoscopy in gastric neoplasms.
Backgrounds: Cycle-consistent generative adversarial network (CycleGAN) is a deep neural network model that performs image-to-image translations. We generated virtual indigo carmine (IC) chromoendoscopy images of gastric neoplasms using CycleGAN and compared their diagnostic performance with that of white light endoscopy (WLE).
Methods: WLE and IC images of 176 patients with gastric neoplasms who underwent endoscopic resection were obtained. We used 1,633 images (911 WLE and 722 IC) of 146 cases in the training dataset to develop virtual IC images using CycleGAN. The remaining 30 WLE images were translated into 30 virtual IC images using the trained CycleGAN and used for validation. The lesion borders were evaluated by 118 endoscopists from 22 institutions using the 60 paired virtual IC and WLE images. The lesion area concordance rate and successful whole-lesion diagnosis were compared.
Results: The lesion area concordance rate based on the pathological diagnosis in virtual IC was lower than in WLE (44.1% vs. 48.5%, p < 0.01). The successful whole-lesion diagnosis was higher in the virtual IC than in WLE images; however, the difference was insignificant (28.2% vs. 26.4%, p = 0.11). Conversely, subgroup analyses revealed a significantly higher diagnosis in virtual IC than in WLE for depressed morphology (41.9% vs. 36.9%, p = 0.02), differentiated histology (27.6% vs. 24.8%, p = 0.02), smaller lesion size (42.3% vs. 38.3%, p = 0.01), and assessed by expert endoscopists (27.3% vs. 23.6%, p = 0.03).
Conclusions: The diagnostic ability of virtual IC was higher for some lesions, but not completely superior to that of WLE. Adjustments are required to improve the imaging system's performance.
期刊介绍:
Gastric Cancer is an esteemed global forum that focuses on various aspects of gastric cancer research, treatment, and biology worldwide.
The journal promotes a diverse range of content, including original articles, case reports, short communications, and technical notes. It also welcomes Letters to the Editor discussing published articles or sharing viewpoints on gastric cancer topics.
Review articles are predominantly sought after by the Editor, ensuring comprehensive coverage of the field.
With a dedicated and knowledgeable editorial team, the journal is committed to providing exceptional support and ensuring high levels of author satisfaction. In fact, over 90% of published authors have expressed their intent to publish again in our esteemed journal.