通过深度学习从多参数磁共振成像中全自动识别子宫内膜癌的淋巴结转移和淋巴管侵犯

IF 4.3 3区材料科学 Q1 ENGINEERING, ELECTRICAL & ELECTRONIC

ACS Applied Electronic Materials Pub Date : 2024-03-12 DOI:10.1002/jmri.29344

Yida Wang MS, Wei Liu MD, Yuanyuan Lu MD, Rennan Ling MD, Wenjing Wang MD, Shengyong Li BS, Feiran Zhang MS, Yan Ning MD, Xiaojun Chen MD, Guang Yang PhD, He Zhang MD

{"title":"通过深度学习从多参数磁共振成像中全自动识别子宫内膜癌的淋巴结转移和淋巴管侵犯","authors":"Yida Wang MS, Wei Liu MD, Yuanyuan Lu MD, Rennan Ling MD, Wenjing Wang MD, Shengyong Li BS, Feiran Zhang MS, Yan Ning MD, Xiaojun Chen MD, Guang Yang PhD, He Zhang MD","doi":"10.1002/jmri.29344","DOIUrl":null,"url":null,"abstract":"<div>\n \n <section>\n \n <h3> Background</h3>\n \n <p>Early and accurate identification of lymphatic node metastasis (LNM) and lymphatic vascular space invasion (LVSI) for endometrial cancer (EC) patients is important for treatment design, but difficult on multi-parametric MRI (mpMRI) images.</p>\n </section>\n \n <section>\n \n <h3> Purpose</h3>\n \n <p>To develop a deep learning (DL) model to simultaneously identify of LNM and LVSI of EC from mpMRI images.</p>\n </section>\n \n <section>\n \n <h3> Study Type</h3>\n \n <p>Retrospective.</p>\n </section>\n \n <section>\n \n <h3> Population</h3>\n \n <p>Six hundred twenty-one patients with histologically proven EC from two institutions, including 111 LNM-positive and 168 LVSI-positive, divided into training, internal, and external test cohorts of 398, 169, and 54 patients, respectively.</p>\n </section>\n \n <section>\n \n <h3> Field Strength/Sequence</h3>\n \n <p>T2-weighted imaging (T2WI), contrast-enhanced T1WI (CE-T1WI), and diffusion-weighted imaging (DWI) were scanned with turbo spin-echo, gradient-echo, and two-dimensional echo-planar sequences, using either a 1.5 T or 3 T system.</p>\n </section>\n \n <section>\n \n <h3> Assessment</h3>\n \n <p>EC lesions were manually delineated on T2WI by two radiologists and used to train an nnU-Net model for automatic segmentation. A multi-task DL model was developed to simultaneously identify LNM and LVSI positive status using the segmented EC lesion regions and T2WI, CE-T1WI, and DWI images as inputs. The performance of the model for LNM-positive diagnosis was compared with those of three radiologists in the external test cohort.</p>\n </section>\n \n <section>\n \n <h3> Statistical Tests</h3>\n \n <p>Dice similarity coefficient (DSC) was used to evaluate segmentation results. Receiver Operating Characteristic (ROC) analysis was used to assess the performance of LNM and LVSI status identification. <i>P</i> value <0.05 was considered significant.</p>\n </section>\n \n <section>\n \n <h3> Results</h3>\n \n <p>EC lesion segmentation model achieved mean DSC values of 0.700 ± 0.25 and 0.693 ± 0.21 in the internal and external test cohorts, respectively. For LNM positive/LVSI positive identification, the proposed model achieved AUC values of 0.895/0.848, 0.806/0.795, and 0.804/0.728 in the training, internal, and external test cohorts, respectively, and better than those of three radiologists (AUC = 0.770/0.648/0.674).</p>\n </section>\n \n <section>\n \n <h3> Data Conclusion</h3>\n \n <p>The proposed model has potential to help clinicians to identify LNM and LVSI status of EC patients and improve treatment planning.</p>\n </section>\n \n <section>\n \n <h3> Evidence Level</h3>\n \n <p>3</p>\n </section>\n \n <section>\n \n <h3> Technical Efficacy</h3>\n \n <p>Stage 2</p>\n </section>\n </div>","PeriodicalId":3,"journal":{"name":"ACS Applied Electronic Materials","volume":null,"pages":null},"PeriodicalIF":4.3000,"publicationDate":"2024-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Fully Automated Identification of Lymph Node Metastases and Lymphovascular Invasion in Endometrial Cancer From Multi-Parametric MRI by Deep Learning\",\"authors\":\"Yida Wang MS, Wei Liu MD, Yuanyuan Lu MD, Rennan Ling MD, Wenjing Wang MD, Shengyong Li BS, Feiran Zhang MS, Yan Ning MD, Xiaojun Chen MD, Guang Yang PhD, He Zhang MD\",\"doi\":\"10.1002/jmri.29344\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div>\\n \\n <section>\\n \\n <h3> Background</h3>\\n \\n <p>Early and accurate identification of lymphatic node metastasis (LNM) and lymphatic vascular space invasion (LVSI) for endometrial cancer (EC) patients is important for treatment design, but difficult on multi-parametric MRI (mpMRI) images.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Purpose</h3>\\n \\n <p>To develop a deep learning (DL) model to simultaneously identify of LNM and LVSI of EC from mpMRI images.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Study Type</h3>\\n \\n <p>Retrospective.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Population</h3>\\n \\n <p>Six hundred twenty-one patients with histologically proven EC from two institutions, including 111 LNM-positive and 168 LVSI-positive, divided into training, internal, and external test cohorts of 398, 169, and 54 patients, respectively.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Field Strength/Sequence</h3>\\n \\n <p>T2-weighted imaging (T2WI), contrast-enhanced T1WI (CE-T1WI), and diffusion-weighted imaging (DWI) were scanned with turbo spin-echo, gradient-echo, and two-dimensional echo-planar sequences, using either a 1.5 T or 3 T system.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Assessment</h3>\\n \\n <p>EC lesions were manually delineated on T2WI by two radiologists and used to train an nnU-Net model for automatic segmentation. A multi-task DL model was developed to simultaneously identify LNM and LVSI positive status using the segmented EC lesion regions and T2WI, CE-T1WI, and DWI images as inputs. The performance of the model for LNM-positive diagnosis was compared with those of three radiologists in the external test cohort.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Statistical Tests</h3>\\n \\n <p>Dice similarity coefficient (DSC) was used to evaluate segmentation results. Receiver Operating Characteristic (ROC) analysis was used to assess the performance of LNM and LVSI status identification. <i>P</i> value <0.05 was considered significant.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Results</h3>\\n \\n <p>EC lesion segmentation model achieved mean DSC values of 0.700 ± 0.25 and 0.693 ± 0.21 in the internal and external test cohorts, respectively. For LNM positive/LVSI positive identification, the proposed model achieved AUC values of 0.895/0.848, 0.806/0.795, and 0.804/0.728 in the training, internal, and external test cohorts, respectively, and better than those of three radiologists (AUC = 0.770/0.648/0.674).</p>\\n </section>\\n \\n <section>\\n \\n <h3> Data Conclusion</h3>\\n \\n <p>The proposed model has potential to help clinicians to identify LNM and LVSI status of EC patients and improve treatment planning.</p>\\n </section>\\n \\n <section>\\n \\n <h3> Evidence Level</h3>\\n \\n <p>3</p>\\n </section>\\n \\n <section>\\n \\n <h3> Technical Efficacy</h3>\\n \\n <p>Stage 2</p>\\n </section>\\n </div>\",\"PeriodicalId\":3,\"journal\":{\"name\":\"ACS Applied Electronic Materials\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":4.3000,\"publicationDate\":\"2024-03-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACS Applied Electronic Materials\",\"FirstCategoryId\":\"3\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1002/jmri.29344\",\"RegionNum\":3,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"ENGINEERING, ELECTRICAL & ELECTRONIC\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACS Applied Electronic Materials","FirstCategoryId":"3","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1002/jmri.29344","RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENGINEERING, ELECTRICAL & ELECTRONIC","Score":null,"Total":0}

引用次数: 0

摘要

背景：目的：开发一种深度学习（DL）模型，从mpMRI图像中同时识别子宫内膜癌（EC）的淋巴结转移（LNM）和淋巴管间隙侵犯（LVSI）：研究对象来自两家机构的621名经组织学证实的EC患者，包括111名LNM阳性患者和168名LVSI阳性患者，分为训练组、内部测试组和外部测试组，分别有398名、169名和54名患者：T2加权成像（T2WI）、对比增强T1WI（CE-T1WI）和弥散加权成像（DWI）采用涡轮自旋回波、梯度回波和二维回波平面序列扫描，使用1.5 T或3 T系统：由两名放射科医生在 T2WI 上手动划分心电图病变，并用它来训练用于自动分割的 nnU-Net 模型。开发了一个多任务 DL 模型，利用分割的心血管病变区域和 T2WI、CE-T1WI 及 DWI 图像作为输入，同时识别 LNM 和 LVSI 阳性状态。该模型在 LNM 阳性诊断中的表现与外部测试队列中三位放射科医生的表现进行了比较：统计测试：使用骰子相似系数（DSC）评估分割结果。接收者操作特征（ROC）分析用于评估 LNM 和 LVSI 状态识别的性能。P 值结果在内部和外部测试组中，EC 病灶分割模型的平均 DSC 值分别为 0.700 ± 0.25 和 0.693 ± 0.21。对于 LNM 阳性/LVSI 阳性的识别，所提出的模型在训练队列、内部队列和外部测试队列中的 AUC 值分别为 0.895/0.848、0.806/0.795 和 0.804/0.728，优于三位放射科医生的 AUC 值（AUC = 0.770/0.648/0.674）：数据结论：所提出的模型有望帮助临床医生识别EC患者的LNM和LVSI状态，并改善治疗计划。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Fully Automated Identification of Lymph Node Metastases and Lymphovascular Invasion in Endometrial Cancer From Multi-Parametric MRI by Deep Learning

Background

Early and accurate identification of lymphatic node metastasis (LNM) and lymphatic vascular space invasion (LVSI) for endometrial cancer (EC) patients is important for treatment design, but difficult on multi-parametric MRI (mpMRI) images.

Purpose

To develop a deep learning (DL) model to simultaneously identify of LNM and LVSI of EC from mpMRI images.

Study Type

Retrospective.

Population

Six hundred twenty-one patients with histologically proven EC from two institutions, including 111 LNM-positive and 168 LVSI-positive, divided into training, internal, and external test cohorts of 398, 169, and 54 patients, respectively.

Field Strength/Sequence

T2-weighted imaging (T2WI), contrast-enhanced T1WI (CE-T1WI), and diffusion-weighted imaging (DWI) were scanned with turbo spin-echo, gradient-echo, and two-dimensional echo-planar sequences, using either a 1.5 T or 3 T system.

Assessment

EC lesions were manually delineated on T2WI by two radiologists and used to train an nnU-Net model for automatic segmentation. A multi-task DL model was developed to simultaneously identify LNM and LVSI positive status using the segmented EC lesion regions and T2WI, CE-T1WI, and DWI images as inputs. The performance of the model for LNM-positive diagnosis was compared with those of three radiologists in the external test cohort.

Statistical Tests

Dice similarity coefficient (DSC) was used to evaluate segmentation results. Receiver Operating Characteristic (ROC) analysis was used to assess the performance of LNM and LVSI status identification. P value <0.05 was considered significant.

Results

EC lesion segmentation model achieved mean DSC values of 0.700 ± 0.25 and 0.693 ± 0.21 in the internal and external test cohorts, respectively. For LNM positive/LVSI positive identification, the proposed model achieved AUC values of 0.895/0.848, 0.806/0.795, and 0.804/0.728 in the training, internal, and external test cohorts, respectively, and better than those of three radiologists (AUC = 0.770/0.648/0.674).