Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study

Abstract Background Prognostic prediction is crucial to guide individual treatment for patients with rectal cancer. We aimed to develop and validated a multitask deep learning model for predicting prognosis in rectal cancer patients. Methods This retrospective study enrolled 321 rectal cancer patien...

Full description

Saved in:
Bibliographic Details
Main Authors: Qiong Ma, Runqi Meng, Ruiting Li, Ling Dai, Fu Shen, Jie Yuan, Danqi Sun, Manman Li, Caixia Fu, Rong Li, Feng Feng, Yonggang Li, Tong Tong, Yajia Gu, Yiqun Sun, Dinggang Shen
Format: Article
Language:English
Published: BMC 2025-06-01
Series:BMC Medical Informatics and Decision Making
Subjects:
Online Access:https://doi.org/10.1186/s12911-025-03050-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850225874555633664
author Qiong Ma
Runqi Meng
Ruiting Li
Ling Dai
Fu Shen
Jie Yuan
Danqi Sun
Manman Li
Caixia Fu
Rong Li
Feng Feng
Yonggang Li
Tong Tong
Yajia Gu
Yiqun Sun
Dinggang Shen
author_facet Qiong Ma
Runqi Meng
Ruiting Li
Ling Dai
Fu Shen
Jie Yuan
Danqi Sun
Manman Li
Caixia Fu
Rong Li
Feng Feng
Yonggang Li
Tong Tong
Yajia Gu
Yiqun Sun
Dinggang Shen
author_sort Qiong Ma
collection DOAJ
description Abstract Background Prognostic prediction is crucial to guide individual treatment for patients with rectal cancer. We aimed to develop and validated a multitask deep learning model for predicting prognosis in rectal cancer patients. Methods This retrospective study enrolled 321 rectal cancer patients (training set: 212; internal testing set: 53; external testing set: 56) who directly received total mesorectal excision from five hospitals between March 2014 to April 2021. A multitask deep learning model was developed to simultaneously predict recurrence/metastasis and disease-free survival (DFS). The model integrated clinicopathologic data and multiparametric magnetic resonance imaging (MRI) images including diffusion kurtosis imaging (DKI), without performing tumor segmentation. The receiver operating characteristic (ROC) curve and Harrell’s concordance index (C-index) were used to evaluate the predictive performance of the proposed model. Results The deep learning model achieved good discrimination capability of recurrence/metastasis, with area under the curve (AUC) values of 0.885, 0.846, and 0.797 in the training, internal testing and external testing sets, respectively. Furthermore, the model successfully predicted DFS in the training set (C-index: 0.812), internal testing set (C-index: 0.794), and external testing set (C-index: 0.733), and classified patients into significantly distinct high- and low-risk groups (p < 0.05). Conclusions The multitask deep learning model, incorporating clinicopathologic data and multiparametric MRI, effectively predicted both recurrence/metastasis and survival for patients with rectal cancer. It has the potential to be an essential tool for risk stratification, and assist in making individualized treatment decisions. Clinical trial number Not applicable.
format Article
id doaj-art-ca3d3f133a474b0c8b9fe8e0aa857098
institution OA Journals
issn 1472-6947
language English
publishDate 2025-06-01
publisher BMC
record_format Article
series BMC Medical Informatics and Decision Making
spelling doaj-art-ca3d3f133a474b0c8b9fe8e0aa8570982025-08-20T02:05:13ZengBMCBMC Medical Informatics and Decision Making1472-69472025-06-0125111510.1186/s12911-025-03050-3Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective studyQiong Ma0Runqi Meng1Ruiting Li2Ling Dai3Fu Shen4Jie Yuan5Danqi Sun6Manman Li7Caixia Fu8Rong Li9Feng Feng10Yonggang Li11Tong Tong12Yajia Gu13Yiqun Sun14Dinggang Shen15Department of Radiology, Fudan University Shanghai Cancer CenterSchool of Biomedical Engineering & State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech UniversityDepartment of Radiology, Fudan University Shanghai Cancer CenterSchool of Biomedical Engineering & State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech UniversityDepartment of Radiology, Changhai Hospital, The Navy Medical UniversityDepartment of Radiology, Shuguang Hospital Affiliated to Shanghai University of Traditional Chinese MedicineDepartment of Radiology, First Affiliated Hospital of Soochow UniversityDepartment of Radiology, Nantong Tumor Hospital, Nantong UniversityMR Application Development, Siemens Shenzhen Magnetic Resonance Ltd.Department of Radiology, Fudan University Shanghai Cancer CenterDepartment of Radiology, Nantong Tumor Hospital, Nantong UniversityDepartment of Radiology, First Affiliated Hospital of Soochow UniversityDepartment of Radiology, Fudan University Shanghai Cancer CenterDepartment of Radiology, Fudan University Shanghai Cancer CenterDepartment of Radiology, Fudan University Shanghai Cancer CenterSchool of Biomedical Engineering & State Key Laboratory of Advanced Medical Materials and Devices, ShanghaiTech UniversityAbstract Background Prognostic prediction is crucial to guide individual treatment for patients with rectal cancer. We aimed to develop and validated a multitask deep learning model for predicting prognosis in rectal cancer patients. Methods This retrospective study enrolled 321 rectal cancer patients (training set: 212; internal testing set: 53; external testing set: 56) who directly received total mesorectal excision from five hospitals between March 2014 to April 2021. A multitask deep learning model was developed to simultaneously predict recurrence/metastasis and disease-free survival (DFS). The model integrated clinicopathologic data and multiparametric magnetic resonance imaging (MRI) images including diffusion kurtosis imaging (DKI), without performing tumor segmentation. The receiver operating characteristic (ROC) curve and Harrell’s concordance index (C-index) were used to evaluate the predictive performance of the proposed model. Results The deep learning model achieved good discrimination capability of recurrence/metastasis, with area under the curve (AUC) values of 0.885, 0.846, and 0.797 in the training, internal testing and external testing sets, respectively. Furthermore, the model successfully predicted DFS in the training set (C-index: 0.812), internal testing set (C-index: 0.794), and external testing set (C-index: 0.733), and classified patients into significantly distinct high- and low-risk groups (p < 0.05). Conclusions The multitask deep learning model, incorporating clinicopathologic data and multiparametric MRI, effectively predicted both recurrence/metastasis and survival for patients with rectal cancer. It has the potential to be an essential tool for risk stratification, and assist in making individualized treatment decisions. Clinical trial number Not applicable.https://doi.org/10.1186/s12911-025-03050-3Deep learningRectal cancerMultimodal dataPrognostic prediction
spellingShingle Qiong Ma
Runqi Meng
Ruiting Li
Ling Dai
Fu Shen
Jie Yuan
Danqi Sun
Manman Li
Caixia Fu
Rong Li
Feng Feng
Yonggang Li
Tong Tong
Yajia Gu
Yiqun Sun
Dinggang Shen
Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
BMC Medical Informatics and Decision Making
Deep learning
Rectal cancer
Multimodal data
Prognostic prediction
title Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
title_full Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
title_fullStr Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
title_full_unstemmed Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
title_short Multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer: a multicenter retrospective study
title_sort multitask deep learning model based on multimodal data for predicting prognosis of rectal cancer a multicenter retrospective study
topic Deep learning
Rectal cancer
Multimodal data
Prognostic prediction
url https://doi.org/10.1186/s12911-025-03050-3
work_keys_str_mv AT qiongma multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT runqimeng multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT ruitingli multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT lingdai multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT fushen multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT jieyuan multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT danqisun multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT manmanli multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT caixiafu multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT rongli multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT fengfeng multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT yonggangli multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT tongtong multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT yajiagu multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT yiqunsun multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy
AT dinggangshen multitaskdeeplearningmodelbasedonmultimodaldataforpredictingprognosisofrectalcanceramulticenterretrospectivestudy