Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer

Abstract Objective To develop and evaluate a intralesional and perilesional radiomics strategy based on different machine learning model to differentiate International Society of Urological Pathology (ISUP) grade > 2 group and ISUP ≤ 2 prostate cancers (PCa). Methods 340 case of PCa patients conf...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhiping Li, Liqin Yang, Ximing Wang, Huijing Xu, Wen Chen, Shuchao Kang, Yasheng Huang, Chang Shu, Feng Cui, Yongsheng Zhang
Format: Article
Language:English
Published: BMC 2025-07-01
Series:BMC Medical Imaging
Subjects:
Online Access:https://doi.org/10.1186/s12880-025-01812-z
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849767254661529600
author Zhiping Li
Liqin Yang
Ximing Wang
Huijing Xu
Wen Chen
Shuchao Kang
Yasheng Huang
Chang Shu
Feng Cui
Yongsheng Zhang
author_facet Zhiping Li
Liqin Yang
Ximing Wang
Huijing Xu
Wen Chen
Shuchao Kang
Yasheng Huang
Chang Shu
Feng Cui
Yongsheng Zhang
author_sort Zhiping Li
collection DOAJ
description Abstract Objective To develop and evaluate a intralesional and perilesional radiomics strategy based on different machine learning model to differentiate International Society of Urological Pathology (ISUP) grade > 2 group and ISUP ≤ 2 prostate cancers (PCa). Methods 340 case of PCa patients confirmed by radical prostatectomy pathology were obtained from two hospitals. The patients were divided into training, internal validation, and external validation groups. Radiomic features were extracted from T2-weighted imaging, and four distinct radiomic feature models were constructed: intralesional, perilesional, combined tumoral and perilesional, and intralesional and perilesional image fusion. Four machine learning classifiers logistic regression (LR), random forest (RF), extra trees (ET), and multilayer perceptron (MLP) were employed for model training and evaluation to select the optimal model. The performance of each model was assessed by calculating the area under the ROC curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. Results The AUCs for the RF classifier were higher than that of LR, ET, and MLP, and was selected as the final radiomic model. The nomogram model integrating perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion had an AUC of 0.929, 0.734, 0.743 for the training, internal, and external validation cohorts, respectively, which was higher than that of the individual intralesional, perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion models. Conclusions The proposed nomogram established from perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion radiomic has the potential to predict the differentiation degree of ISUP PCa patients. Clinical trial number Not applicable.
format Article
id doaj-art-e55e552afa594bf4864b918a48b005a7
institution DOAJ
issn 1471-2342
language English
publishDate 2025-07-01
publisher BMC
record_format Article
series BMC Medical Imaging
spelling doaj-art-e55e552afa594bf4864b918a48b005a72025-08-20T03:04:17ZengBMCBMC Medical Imaging1471-23422025-07-0125111410.1186/s12880-025-01812-zIntralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancerZhiping Li0Liqin Yang1Ximing Wang2Huijing Xu3Wen Chen4Shuchao Kang5Yasheng Huang6Chang Shu7Feng Cui8Yongsheng Zhang9Department of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, The First Affiliated Hospital of Soochow UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Urology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Pathology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityDepartment of Radiology, Hangzhou TCM Hospital Affiliated to Zhejiang Chinese Medical UniversityAbstract Objective To develop and evaluate a intralesional and perilesional radiomics strategy based on different machine learning model to differentiate International Society of Urological Pathology (ISUP) grade > 2 group and ISUP ≤ 2 prostate cancers (PCa). Methods 340 case of PCa patients confirmed by radical prostatectomy pathology were obtained from two hospitals. The patients were divided into training, internal validation, and external validation groups. Radiomic features were extracted from T2-weighted imaging, and four distinct radiomic feature models were constructed: intralesional, perilesional, combined tumoral and perilesional, and intralesional and perilesional image fusion. Four machine learning classifiers logistic regression (LR), random forest (RF), extra trees (ET), and multilayer perceptron (MLP) were employed for model training and evaluation to select the optimal model. The performance of each model was assessed by calculating the area under the ROC curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. Results The AUCs for the RF classifier were higher than that of LR, ET, and MLP, and was selected as the final radiomic model. The nomogram model integrating perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion had an AUC of 0.929, 0.734, 0.743 for the training, internal, and external validation cohorts, respectively, which was higher than that of the individual intralesional, perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion models. Conclusions The proposed nomogram established from perilesional, combined intralesional and perilesional, and intralesional and perilesional image fusion radiomic has the potential to predict the differentiation degree of ISUP PCa patients. Clinical trial number Not applicable.https://doi.org/10.1186/s12880-025-01812-zMagnetic resonance imagingPerilesional radiomicsMachine learningProstate cancerInternational society of urological pathology grading
spellingShingle Zhiping Li
Liqin Yang
Ximing Wang
Huijing Xu
Wen Chen
Shuchao Kang
Yasheng Huang
Chang Shu
Feng Cui
Yongsheng Zhang
Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
BMC Medical Imaging
Magnetic resonance imaging
Perilesional radiomics
Machine learning
Prostate cancer
International society of urological pathology grading
title Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
title_full Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
title_fullStr Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
title_full_unstemmed Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
title_short Intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
title_sort intralesional and perilesional radiomics strategy based on different machine learning for the prediction of international society of urological pathology grade group in prostate cancer
topic Magnetic resonance imaging
Perilesional radiomics
Machine learning
Prostate cancer
International society of urological pathology grading
url https://doi.org/10.1186/s12880-025-01812-z
work_keys_str_mv AT zhipingli intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT liqinyang intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT ximingwang intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT huijingxu intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT wenchen intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT shuchaokang intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT yashenghuang intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT changshu intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT fengcui intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer
AT yongshengzhang intralesionalandperilesionalradiomicsstrategybasedondifferentmachinelearningforthepredictionofinternationalsocietyofurologicalpathologygradegroupinprostatecancer