A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model

PurposeThis study aimed to establish and evaluate a model utilizing bi-parametric ultrasound-based deep learning radiomics (DLR) in conjunction with clinical factors to anticipate clinically significant prostate cancer (csPCa).MethodsWe retrospectively analyzed 232 participants from our institution...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiang Liu, Zhong-Xin Zhang, Bing Zheng, Min Xu, Xin-Yu Cao, Hai-Ming Huang
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-04-01
Series:Frontiers in Oncology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fonc.2025.1538854/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702119372750848
author Xiang Liu
Zhong-Xin Zhang
Bing Zheng
Min Xu
Xin-Yu Cao
Hai-Ming Huang
author_facet Xiang Liu
Zhong-Xin Zhang
Bing Zheng
Min Xu
Xin-Yu Cao
Hai-Ming Huang
author_sort Xiang Liu
collection DOAJ
description PurposeThis study aimed to establish and evaluate a model utilizing bi-parametric ultrasound-based deep learning radiomics (DLR) in conjunction with clinical factors to anticipate clinically significant prostate cancer (csPCa).MethodsWe retrospectively analyzed 232 participants from our institution who underwent both B-mode ultrasound and shear wave elastography (SWE) prior to prostate biopsy between June 2022 and December 2023. A random allocation placed the participants into training and test cohorts with a 7:3 distribution. We developed a nomogram that integrates DLR with clinical factors within the training cohort, which was subsequently validated using the test cohort. The diagnostic performance and clinical applicability were evaluated with receiver operating characteristic (ROC) curve analysis and decision curve analysis.ResultsIn our study, the bi-parametric ultrasound-based DLR model demonstrated an area under the curve (AUC) of 0.80 (95%CI: 0.70-0.91) in the test set, surpassing the performance of both the radiomics and deep learning models individually. By integrating clinical factors, a composite model, presented as the nomogram, was developed and exhibited superior diagnostic performance, achieving an AUC of 0.87 (95%CI: 0.77-0.95) in the test set. The performance exceeded that of the DLR (P = 0.049) and the clinical model (AUC = 0.79, 95%CI: 0.69-0.86, P = 0.041). Furthermore, the decision curve analysis indicated that the composite model provided a greater net benefit across a various high-risk threshold than the DLR or the clinical model alone.ConclusionTo our knowledge, this is the first proposal of a nomogram integrating ultrasound-based DLR with clinical indicators for predicting csPCa. This nomogram can improve the accuracy of csPCa prediction and may help physicians make more confident decisions regarding interventions, particularly in settings where MRI is unavailable.
format Article
id doaj-art-a6422cddfa5d48808f29d9ac26ed7aa4
institution DOAJ
issn 2234-943X
language English
publishDate 2025-04-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Oncology
spelling doaj-art-a6422cddfa5d48808f29d9ac26ed7aa42025-08-20T03:17:44ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-04-011510.3389/fonc.2025.15388541538854A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics modelXiang Liu0Zhong-Xin Zhang1Bing Zheng2Min Xu3Xin-Yu Cao4Hai-Ming Huang5Department of Ultrasound, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Urology Surgery, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaDepartment of Ultrasound, The Second Affiliated Hospital of Nantong University, Nantong, Jiangsu, ChinaPurposeThis study aimed to establish and evaluate a model utilizing bi-parametric ultrasound-based deep learning radiomics (DLR) in conjunction with clinical factors to anticipate clinically significant prostate cancer (csPCa).MethodsWe retrospectively analyzed 232 participants from our institution who underwent both B-mode ultrasound and shear wave elastography (SWE) prior to prostate biopsy between June 2022 and December 2023. A random allocation placed the participants into training and test cohorts with a 7:3 distribution. We developed a nomogram that integrates DLR with clinical factors within the training cohort, which was subsequently validated using the test cohort. The diagnostic performance and clinical applicability were evaluated with receiver operating characteristic (ROC) curve analysis and decision curve analysis.ResultsIn our study, the bi-parametric ultrasound-based DLR model demonstrated an area under the curve (AUC) of 0.80 (95%CI: 0.70-0.91) in the test set, surpassing the performance of both the radiomics and deep learning models individually. By integrating clinical factors, a composite model, presented as the nomogram, was developed and exhibited superior diagnostic performance, achieving an AUC of 0.87 (95%CI: 0.77-0.95) in the test set. The performance exceeded that of the DLR (P = 0.049) and the clinical model (AUC = 0.79, 95%CI: 0.69-0.86, P = 0.041). Furthermore, the decision curve analysis indicated that the composite model provided a greater net benefit across a various high-risk threshold than the DLR or the clinical model alone.ConclusionTo our knowledge, this is the first proposal of a nomogram integrating ultrasound-based DLR with clinical indicators for predicting csPCa. This nomogram can improve the accuracy of csPCa prediction and may help physicians make more confident decisions regarding interventions, particularly in settings where MRI is unavailable.https://www.frontiersin.org/articles/10.3389/fonc.2025.1538854/fullbi-parametricultrasounddeep learningradiomicsprostate cancer
spellingShingle Xiang Liu
Zhong-Xin Zhang
Bing Zheng
Min Xu
Xin-Yu Cao
Hai-Ming Huang
A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
Frontiers in Oncology
bi-parametric
ultrasound
deep learning
radiomics
prostate cancer
title A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
title_full A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
title_fullStr A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
title_full_unstemmed A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
title_short A retrospective study on predicting clinically significant prostate cancer via a bi-parametric ultrasound-based deep learning radiomics model
title_sort retrospective study on predicting clinically significant prostate cancer via a bi parametric ultrasound based deep learning radiomics model
topic bi-parametric
ultrasound
deep learning
radiomics
prostate cancer
url https://www.frontiersin.org/articles/10.3389/fonc.2025.1538854/full
work_keys_str_mv AT xiangliu aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT zhongxinzhang aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT bingzheng aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT minxu aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT xinyucao aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT haiminghuang aretrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT xiangliu retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT zhongxinzhang retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT bingzheng retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT minxu retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT xinyucao retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel
AT haiminghuang retrospectivestudyonpredictingclinicallysignificantprostatecancerviaabiparametricultrasoundbaseddeeplearningradiomicsmodel