AutoLNMNet: Automated Network for Estimating Lymph-Node Metastasis in EGC Using a Pyramid Vision Transformer and Data Derived From Multiphoton Microscopy.

IF 2 3区工程技术 Q2 ANATOMY & MORPHOLOGY Microscopy Research and Technique Pub Date : 2024-10-01 DOI:10.1002/jemt.24705

Lin Gao, Wenju Liu, Bingzi Kang, Han Wu, Jiajia He, Xiaolu Li, Gangqin Xi, Shuangmu Zhuo

{"title":"AutoLNMNet: Automated Network for Estimating Lymph-Node Metastasis in EGC Using a Pyramid Vision Transformer and Data Derived From Multiphoton Microscopy.","authors":"Lin Gao, Wenju Liu, Bingzi Kang, Han Wu, Jiajia He, Xiaolu Li, Gangqin Xi, Shuangmu Zhuo","doi":"10.1002/jemt.24705","DOIUrl":null,"url":null,"abstract":"<p><p>Lymph-node status is important in decision-making during early gastric cancer (EGC) treatment. Currently, endoscopic submucosal dissection is the mainstream treatment for EGC. However, it is challenging for even experienced endoscopists to accurately diagnose and treat EGC. Multiphoton microscopy can extract the morphological features of collagen fibers from tissues. The characteristics of collagen fibers can be used to assess the lymph-node metastasis status in patients with EGC. First, we compared the accuracy of four deep learning models (VGG16, ResNet34, MobileNetV2, and PVTv2) in training preprocessed images and test datasets. Next, we integrated the features of the best-performing model, which was PVTv2, with manual and clinical features to develop a novel model called AutoLNMNet. The prediction accuracy of AutoLNMNet for the no metastasis (Ly0) and metastasis in lymph nodes (Ly1) stages reached 0.92, which was 0.3% higher than that of PVTv2. The receiver operating characteristics of AutoLNMNet in quantifying Ly0 and Ly1 stages were 0.97 and 0.97, respectively. Therefore, AutoLNMNet is highly reliable and accurate in detecting lymph-node metastasis, providing an important tool for the early diagnosis and treatment of EGC.</p>","PeriodicalId":18684,"journal":{"name":"Microscopy Research and Technique","volume":null,"pages":null},"PeriodicalIF":2.0000,"publicationDate":"2024-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Microscopy Research and Technique","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1002/jemt.24705","RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ANATOMY & MORPHOLOGY","Score":null,"Total":0}

引用次数: 0

Abstract

Lymph-node status is important in decision-making during early gastric cancer (EGC) treatment. Currently, endoscopic submucosal dissection is the mainstream treatment for EGC. However, it is challenging for even experienced endoscopists to accurately diagnose and treat EGC. Multiphoton microscopy can extract the morphological features of collagen fibers from tissues. The characteristics of collagen fibers can be used to assess the lymph-node metastasis status in patients with EGC. First, we compared the accuracy of four deep learning models (VGG16, ResNet34, MobileNetV2, and PVTv2) in training preprocessed images and test datasets. Next, we integrated the features of the best-performing model, which was PVTv2, with manual and clinical features to develop a novel model called AutoLNMNet. The prediction accuracy of AutoLNMNet for the no metastasis (Ly0) and metastasis in lymph nodes (Ly1) stages reached 0.92, which was 0.3% higher than that of PVTv2. The receiver operating characteristics of AutoLNMNet in quantifying Ly0 and Ly1 stages were 0.97 and 0.97, respectively. Therefore, AutoLNMNet is highly reliable and accurate in detecting lymph-node metastasis, providing an important tool for the early diagnosis and treatment of EGC.

查看原文

微信好友朋友圈 QQ好友复制链接

本刊更多论文

AutoLNMNet：使用金字塔视觉转换器和多光子显微镜数据估算 EGC 淋巴结转移的自动网络

淋巴结状态是早期胃癌（EGC）治疗决策的重要依据。目前，内镜粘膜下剥离术是治疗 EGC 的主流方法。然而，即使是经验丰富的内镜医师，要准确诊断和治疗 EGC 也是一项挑战。多光子显微镜可以从组织中提取胶原纤维的形态特征。胶原纤维的特征可用于评估EGC患者的淋巴结转移状况。首先，我们比较了四种深度学习模型（VGG16、ResNet34、MobileNetV2 和 PVTv2）在训练预处理图像和测试数据集中的准确性。接下来，我们将表现最好的模型（PVTv2）的特征与人工特征和临床特征相结合，开发出一种名为 AutoLNMNet 的新型模型。AutoLNMNet 对无转移期（Ly0）和淋巴结转移期（Ly1）的预测准确率达到 0.92，比 PVTv2 高 0.3%。AutoLNMNet 定量 Ly0 和 Ly1 分期的接收器操作特性分别为 0.97 和 0.97。因此，AutoLNMNet 在检测淋巴结转移方面具有高度可靠性和准确性，为 EGC 的早期诊断和治疗提供了重要工具。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文去求助

来源期刊

Microscopy Research and Technique 医学-解剖学与形态学

CiteScore

5.30

自引率

20.00%

发文量

233

审稿时长

4.7 months

期刊介绍： Microscopy Research and Technique (MRT) publishes articles on all aspects of advanced microscopy original architecture and methodologies with applications in the biological, clinical, chemical, and materials sciences. Original basic and applied research as well as technical papers dealing with the various subsets of microscopy are encouraged. MRT is the right form for those developing new microscopy methods or using the microscope to answer key questions in basic and applied research.