Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study

Abstract To develop a deep learning (DL) model based on MRI to predict muscle-invasive bladder cancer (MIBC). A total of 559 patients, including 521 patients in our center and 38 patients in external centers were collected from 2012 to 2023 to construct the DL model. In this study, the DL model was...

Full description

Saved in:
Bibliographic Details
Main Authors: Lingkai Cai, Xiao Yang, Jie Yu, Qiang Shao, Gongcheng Wang, Baorui Yuan, Juntao Zhuang, Kai Li, Qikai Wu, Peikun Liu, Ruixi Yu, Qiang Cao, Pengchao Li, Xueying Sun, Yuan Zou, Xue Fu, Xiangming Fang, Chunxiao Chen, Qiang Lu
Format: Article
Language:English
Published: Nature Portfolio 2025-03-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-024-82909-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850094842463387648
author Lingkai Cai
Xiao Yang
Jie Yu
Qiang Shao
Gongcheng Wang
Baorui Yuan
Juntao Zhuang
Kai Li
Qikai Wu
Peikun Liu
Ruixi Yu
Qiang Cao
Pengchao Li
Xueying Sun
Yuan Zou
Xue Fu
Xiangming Fang
Chunxiao Chen
Qiang Lu
author_facet Lingkai Cai
Xiao Yang
Jie Yu
Qiang Shao
Gongcheng Wang
Baorui Yuan
Juntao Zhuang
Kai Li
Qikai Wu
Peikun Liu
Ruixi Yu
Qiang Cao
Pengchao Li
Xueying Sun
Yuan Zou
Xue Fu
Xiangming Fang
Chunxiao Chen
Qiang Lu
author_sort Lingkai Cai
collection DOAJ
description Abstract To develop a deep learning (DL) model based on MRI to predict muscle-invasive bladder cancer (MIBC). A total of 559 patients, including 521 patients in our center and 38 patients in external centers were collected from 2012 to 2023 to construct the DL model. In this study, the DL model was utilized to differentiate between MIBC and NMIBC based on three-channel image inputs, including original T2WI images, segmented bladder, and regions of interest. Inception V3 was employed for model construction. The accuracy, sensitivity (SN), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) for predicting MIBC by DL model were 92.4%, 94.7%, 91.5%, 81.8% and 97.7% in the validation set and 92.1%, 86.8%, 94.6%, 88.5% and 93.8% in the internal test set. In the external test set, these values were 81.6%, 57.1%, 87.1%, 50.0% and 90.0%. Additionally, the accuracy, SN, SP, PPV, and NPV for predicting MIBC were 93.5%, 100%, 93.4%, 11.1%, and 100% in VI-RADS 2; 80.0%, 66.7%, 87.2%, 73.7% and 82.9% in VI-RADS 3; 90.3%, 91.7%, 85.7%, 95.7%, 75.0% in VI-RADS 4. The accuracy, SN, and PPV were 93.9%, 93.9%, and 100% in VI-RADS 5. The DL model based on T2WI can effectively predict MIBC and serve as a valuable complement to VI-RADS 3.
format Article
id doaj-art-68e11aae5a334c06abea6f1e84c7ad4c
institution DOAJ
issn 2045-2322
language English
publishDate 2025-03-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-68e11aae5a334c06abea6f1e84c7ad4c2025-08-20T02:41:34ZengNature PortfolioScientific Reports2045-23222025-03-011511910.1038/s41598-024-82909-3Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical studyLingkai Cai0Xiao Yang1Jie Yu2Qiang Shao3Gongcheng Wang4Baorui Yuan5Juntao Zhuang6Kai Li7Qikai Wu8Peikun Liu9Ruixi Yu10Qiang Cao11Pengchao Li12Xueying Sun13Yuan Zou14Xue Fu15Xiangming Fang16Chunxiao Chen17Qiang Lu18Department of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Biomedical Engineering, Nanjing University of Aeronautics and AstronauticsDepartment of Urology, Suzhou Hospital Affiliated of Nanjing Medical UniversityDepartment of Urology, The Affiliated Huaian No.1 People’s Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Radiology, The First Affiliated Hospital of Nanjing Medical UniversityDepartment of Biomedical Engineering, Nanjing University of Aeronautics and AstronauticsDepartment of Biomedical Engineering, Nanjing University of Aeronautics and AstronauticsDepartment of Radiology, Wuxi Medical Center of Nanjing Medical UniversityDepartment of Biomedical Engineering, Nanjing University of Aeronautics and AstronauticsDepartment of Urology, The First Affiliated Hospital of Nanjing Medical UniversityAbstract To develop a deep learning (DL) model based on MRI to predict muscle-invasive bladder cancer (MIBC). A total of 559 patients, including 521 patients in our center and 38 patients in external centers were collected from 2012 to 2023 to construct the DL model. In this study, the DL model was utilized to differentiate between MIBC and NMIBC based on three-channel image inputs, including original T2WI images, segmented bladder, and regions of interest. Inception V3 was employed for model construction. The accuracy, sensitivity (SN), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV) for predicting MIBC by DL model were 92.4%, 94.7%, 91.5%, 81.8% and 97.7% in the validation set and 92.1%, 86.8%, 94.6%, 88.5% and 93.8% in the internal test set. In the external test set, these values were 81.6%, 57.1%, 87.1%, 50.0% and 90.0%. Additionally, the accuracy, SN, SP, PPV, and NPV for predicting MIBC were 93.5%, 100%, 93.4%, 11.1%, and 100% in VI-RADS 2; 80.0%, 66.7%, 87.2%, 73.7% and 82.9% in VI-RADS 3; 90.3%, 91.7%, 85.7%, 95.7%, 75.0% in VI-RADS 4. The accuracy, SN, and PPV were 93.9%, 93.9%, and 100% in VI-RADS 5. The DL model based on T2WI can effectively predict MIBC and serve as a valuable complement to VI-RADS 3.https://doi.org/10.1038/s41598-024-82909-3Bladder cancerMRIMIBCDeep learning
spellingShingle Lingkai Cai
Xiao Yang
Jie Yu
Qiang Shao
Gongcheng Wang
Baorui Yuan
Juntao Zhuang
Kai Li
Qikai Wu
Peikun Liu
Ruixi Yu
Qiang Cao
Pengchao Li
Xueying Sun
Yuan Zou
Xue Fu
Xiangming Fang
Chunxiao Chen
Qiang Lu
Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
Scientific Reports
Bladder cancer
MRI
MIBC
Deep learning
title Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
title_full Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
title_fullStr Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
title_full_unstemmed Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
title_short Deep learning on T2WI to predict the muscle-invasive bladder cancer: a multi-center clinical study
title_sort deep learning on t2wi to predict the muscle invasive bladder cancer a multi center clinical study
topic Bladder cancer
MRI
MIBC
Deep learning
url https://doi.org/10.1038/s41598-024-82909-3
work_keys_str_mv AT lingkaicai deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT xiaoyang deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT jieyu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT qiangshao deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT gongchengwang deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT baoruiyuan deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT juntaozhuang deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT kaili deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT qikaiwu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT peikunliu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT ruixiyu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT qiangcao deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT pengchaoli deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT xueyingsun deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT yuanzou deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT xuefu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT xiangmingfang deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT chunxiaochen deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy
AT qianglu deeplearningont2witopredictthemuscleinvasivebladdercanceramulticenterclinicalstudy