Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning.

IF 5 Q1 ENGINEERING, BIOMEDICAL BME frontiers Pub Date : 2022-03-16 eCollection Date: 2022-01-01 DOI:10.34133/2022/9860179
Hailing Liu, Yu Zhao, Fan Yang, Xiaoying Lou, Feng Wu, Hang Li, Xiaohan Xing, Tingying Peng, Bjoern Menze, Junzhou Huang, Shujun Zhang, Anjia Han, Jianhua Yao, Xinjuan Fan
{"title":"Preoperative Prediction of Lymph Node Metastasis in Colorectal Cancer with Deep Learning.","authors":"Hailing Liu,&nbsp;Yu Zhao,&nbsp;Fan Yang,&nbsp;Xiaoying Lou,&nbsp;Feng Wu,&nbsp;Hang Li,&nbsp;Xiaohan Xing,&nbsp;Tingying Peng,&nbsp;Bjoern Menze,&nbsp;Junzhou Huang,&nbsp;Shujun Zhang,&nbsp;Anjia Han,&nbsp;Jianhua Yao,&nbsp;Xinjuan Fan","doi":"10.34133/2022/9860179","DOIUrl":null,"url":null,"abstract":"<p><p><i>Objective</i>. To develop an artificial intelligence method predicting lymph node metastasis (LNM) for patients with colorectal cancer (CRC). <i>Impact Statement</i>. A novel interpretable multimodal AI-based method to predict LNM for CRC patients by integrating information of pathological images and serum tumor-specific biomarkers. <i>Introduction</i>. Preoperative diagnosis of LNM is essential in treatment planning for CRC patients. Existing radiology imaging and genomic tests approaches are either unreliable or too costly. <i>Methods</i>. A total of 1338 patients were recruited, where 1128 patients from one centre were included as the discovery cohort and 210 patients from other two centres were involved as the external validation cohort. We developed a Multimodal Multiple Instance Learning (MMIL) model to learn latent features from pathological images and then jointly integrated the clinical biomarker features for predicting LNM status. The heatmaps of the obtained MMIL model were generated for model interpretation. <i>Results</i>. The MMIL model outperformed preoperative radiology-imaging diagnosis and yielded high area under the curve (AUCs) of 0.926, 0.878, 0.809, and 0.857 for patients with stage T1, T2, T3, and T4 CRC, on the discovery cohort. On the external cohort, it obtained AUCs of 0.855, 0.832, 0.691, and 0.792, respectively (T1-T4), which indicates its prediction accuracy and potential adaptability among multiple centres. <i>Conclusion</i>. The MMIL model showed the potential in the early diagnosis of LNM by referring to pathological images and tumor-specific biomarkers, which is easily accessed in different institutes. We revealed the histomorphologic features determining the LNM prediction indicating the model ability to learn informative latent features.</p>","PeriodicalId":72430,"journal":{"name":"BME frontiers","volume":"2022 ","pages":"9860179"},"PeriodicalIF":5.0000,"publicationDate":"2022-03-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10521754/pdf/","citationCount":"6","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"BME frontiers","FirstCategoryId":"1087","ListUrlMain":"https://doi.org/10.34133/2022/9860179","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2022/1/1 0:00:00","PubModel":"eCollection","JCR":"Q1","JCRName":"ENGINEERING, BIOMEDICAL","Score":null,"Total":0}
引用次数: 6

Abstract

Objective. To develop an artificial intelligence method predicting lymph node metastasis (LNM) for patients with colorectal cancer (CRC). Impact Statement. A novel interpretable multimodal AI-based method to predict LNM for CRC patients by integrating information of pathological images and serum tumor-specific biomarkers. Introduction. Preoperative diagnosis of LNM is essential in treatment planning for CRC patients. Existing radiology imaging and genomic tests approaches are either unreliable or too costly. Methods. A total of 1338 patients were recruited, where 1128 patients from one centre were included as the discovery cohort and 210 patients from other two centres were involved as the external validation cohort. We developed a Multimodal Multiple Instance Learning (MMIL) model to learn latent features from pathological images and then jointly integrated the clinical biomarker features for predicting LNM status. The heatmaps of the obtained MMIL model were generated for model interpretation. Results. The MMIL model outperformed preoperative radiology-imaging diagnosis and yielded high area under the curve (AUCs) of 0.926, 0.878, 0.809, and 0.857 for patients with stage T1, T2, T3, and T4 CRC, on the discovery cohort. On the external cohort, it obtained AUCs of 0.855, 0.832, 0.691, and 0.792, respectively (T1-T4), which indicates its prediction accuracy and potential adaptability among multiple centres. Conclusion. The MMIL model showed the potential in the early diagnosis of LNM by referring to pathological images and tumor-specific biomarkers, which is easily accessed in different institutes. We revealed the histomorphologic features determining the LNM prediction indicating the model ability to learn informative latent features.

Abstract Image

Abstract Image

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
癌症结直肠癌术前淋巴结转移的深度学习预测。
客观的开发一种预测癌症(CRC)患者淋巴结转移(LNM)的人工智能方法。影响声明。一种新的可解释的基于多模式AI的方法,通过整合病理图像和血清肿瘤特异性生物标志物的信息来预测CRC患者的LNM。介绍LNM的术前诊断对CRC患者的治疗计划至关重要。现有的放射学成像和基因组测试方法要么不可靠,要么成本太高。方法。共招募了1338名患者,其中来自一个中心的1128名患者被纳入发现队列,来自其他两个中心的210名患者被参与外部验证队列。我们开发了一个多模式多实例学习(MMIL)模型来从病理图像中学习潜在特征,然后联合整合临床生物标志物特征来预测LNM状态。生成所获得的MMIL模型的热图用于模型解释。后果MMIL模型优于术前放射学成像诊断,在发现队列中,T1、T2、T3和T4期CRC患者的曲线下面积(AUCs)分别为0.926、0.878、0.809和0.857。在外部队列中,它获得的AUC分别为0.855、0.832、0.691和0.792(T1-T4),这表明它的预测准确性和在多个中心之间的潜在适应性。结论MMIL模型通过参考病理图像和肿瘤特异性生物标志物,显示了在LNM早期诊断中的潜力,这在不同的研究所很容易获得。我们揭示了决定LNM预测的组织形态学特征,表明模型学习信息性潜在特征的能力。
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
CiteScore
7.10
自引率
0.00%
发文量
0
审稿时长
16 weeks
期刊最新文献
A Novel System for Fabricating Microspheres with Microelectromechanical System-Based Bioprinting Technology. A Janus Adhesive Hydrogel with Integrated Attack and Defense for Bacteria Killing and Antifouling. Cationized Decalcified Bone Matrix for Infected Bone Defect Treatment. Functional Neural Networks in Human Brain Organoids. What Is the Magical Cavitation Bubble: A Holistic Perspective to Trigger Advanced Bubbles, Nano-Sonocatalysts, and Cellular Sonosensitizers.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1