Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models

ObjectiveProstate cancer is prevalent among older men. Although this malignancy has a relatively low mortality rate, its aggressiveness is critical in determining patient prognosis and treatment options. This study therefore aimed to evaluate the effectiveness of a 2.5D deep learning model based on...

Full description

Saved in:
Bibliographic Details
Main Authors: Yalei Wang, Yuqing Xin, Baoqi Zhang, Fuqiang Pan, Xu Li, Manman Zhang, Yushan Yuan, Lei Zhang, Peiqi Ma, Bo Guan, Yang Zhang
Format: Article
Language:English
Published: Frontiers Media S.A. 2025-06-01
Series:Frontiers in Oncology
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/fonc.2025.1539537/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850111049275015168
author Yalei Wang
Yuqing Xin
Baoqi Zhang
Fuqiang Pan
Xu Li
Manman Zhang
Yushan Yuan
Lei Zhang
Peiqi Ma
Bo Guan
Yang Zhang
author_facet Yalei Wang
Yuqing Xin
Baoqi Zhang
Fuqiang Pan
Xu Li
Manman Zhang
Yushan Yuan
Lei Zhang
Peiqi Ma
Bo Guan
Yang Zhang
author_sort Yalei Wang
collection DOAJ
description ObjectiveProstate cancer is prevalent among older men. Although this malignancy has a relatively low mortality rate, its aggressiveness is critical in determining patient prognosis and treatment options. This study therefore aimed to evaluate the effectiveness of a 2.5D deep learning model based on prostate MRI to assess prostate cancer aggressiveness.Materials and methodsThis study included 335 patients with pathologically-confirmed prostate cancer from a tertiary medical center between January 2022 and December 2023. Of these, 266 cases were classified as aggressive and 69 as non-aggressive, using a Gleason score ≥7 as the cutoff. The subjects were automatically divided into a test set and validation set in a 7:3 ratio. Before pathological biopsy, all patients underwent biparametric MRI, including T2-weighted imaging, diffusion-weighted imaging, and apparent diffusion coefficient scans. Two radiologists, blinded to pathology results, segmented the lesions using ITK-SNAP software, extracting the minimal bounding rectangle of the largest ROI layer, along with the corresponding ROIs from adjacent layers above and below it. Subsequently, radiomic features were extracted using pyradiomics tool, while deep learning features from each cross-section were derived using the Inception_v3 neural network. To ensure consistency in feature extraction, intraclass correlation coefficient (ICC) analysis was performed on features extracted by radiologists, followed by feature normalization using the mean and standard deviation of the training set. Highly correlated features were removed using t-tests and Pearson correlation tests, and redundant features were ultimately screened with least absolute shrinkage and selection operator (Lasso). Models were constructed using the LightGBM algorithm: a radiomic feature model, a deep learning feature model, and a combined model integrating radiomic and deep learning features. Further, a clinical feature model (Clinic-LightGBM) was constructed using LightGBM to include clinical information. The optimal feature model was then combined with Clinic-LightGBM to establish a nomogram. The Grad-CAM technique was employed to explain the deep learning feature extraction process, supported by tree model visualization techniques to illustrate the decision-making process of the LightGBM model. Model classification performance in the test set was evaluated using the area under the receiver operating characteristic curve (AUC).ResultsIn the test set, the nomogram demonstrated the highest predictive ability for prostate cancer aggressiveness (AUC = 0.919, 95% CI: 0.8107–1.0000), with a sensitivity of 0.966 and specificity of 0.833. The DLR-LightGBM model (AUC = 0.872) outperformed the DL-LightGBM (AUC = 0.818) and Rad-LightGBM (AUC = 0.758) models, indicating the benefit of combining deep learning and radiomic features.ConclusionOur 2.5D deep learning model based on prostate MRI showed efficacy in identifying clinically significant prostate cancer, providing valuable references for clinical treatment and enhancing patient net benefit.
format Article
id doaj-art-9804d31c252f4ad092ed7b963bb21386
institution OA Journals
issn 2234-943X
language English
publishDate 2025-06-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Oncology
spelling doaj-art-9804d31c252f4ad092ed7b963bb213862025-08-20T02:37:42ZengFrontiers Media S.A.Frontiers in Oncology2234-943X2025-06-011510.3389/fonc.2025.15395371539537Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning modelsYalei Wang0Yuqing Xin1Baoqi Zhang2Fuqiang Pan3Xu Li4Manman Zhang5Yushan Yuan6Lei Zhang7Peiqi Ma8Bo Guan9Yang Zhang10Department of Radiology, Fuyang People’s Hospital of Anhui Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Bengbu Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Anhui Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Bengbu Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Bengbu Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Bengbu Medical University, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital, Fuyang, ChinaDepartment of Urology, Fuyang People’s Hospital, Fuyang, ChinaDepartment of Radiology, Fuyang People’s Hospital of Anhui Medical University, Fuyang, ChinaObjectiveProstate cancer is prevalent among older men. Although this malignancy has a relatively low mortality rate, its aggressiveness is critical in determining patient prognosis and treatment options. This study therefore aimed to evaluate the effectiveness of a 2.5D deep learning model based on prostate MRI to assess prostate cancer aggressiveness.Materials and methodsThis study included 335 patients with pathologically-confirmed prostate cancer from a tertiary medical center between January 2022 and December 2023. Of these, 266 cases were classified as aggressive and 69 as non-aggressive, using a Gleason score ≥7 as the cutoff. The subjects were automatically divided into a test set and validation set in a 7:3 ratio. Before pathological biopsy, all patients underwent biparametric MRI, including T2-weighted imaging, diffusion-weighted imaging, and apparent diffusion coefficient scans. Two radiologists, blinded to pathology results, segmented the lesions using ITK-SNAP software, extracting the minimal bounding rectangle of the largest ROI layer, along with the corresponding ROIs from adjacent layers above and below it. Subsequently, radiomic features were extracted using pyradiomics tool, while deep learning features from each cross-section were derived using the Inception_v3 neural network. To ensure consistency in feature extraction, intraclass correlation coefficient (ICC) analysis was performed on features extracted by radiologists, followed by feature normalization using the mean and standard deviation of the training set. Highly correlated features were removed using t-tests and Pearson correlation tests, and redundant features were ultimately screened with least absolute shrinkage and selection operator (Lasso). Models were constructed using the LightGBM algorithm: a radiomic feature model, a deep learning feature model, and a combined model integrating radiomic and deep learning features. Further, a clinical feature model (Clinic-LightGBM) was constructed using LightGBM to include clinical information. The optimal feature model was then combined with Clinic-LightGBM to establish a nomogram. The Grad-CAM technique was employed to explain the deep learning feature extraction process, supported by tree model visualization techniques to illustrate the decision-making process of the LightGBM model. Model classification performance in the test set was evaluated using the area under the receiver operating characteristic curve (AUC).ResultsIn the test set, the nomogram demonstrated the highest predictive ability for prostate cancer aggressiveness (AUC = 0.919, 95% CI: 0.8107–1.0000), with a sensitivity of 0.966 and specificity of 0.833. The DLR-LightGBM model (AUC = 0.872) outperformed the DL-LightGBM (AUC = 0.818) and Rad-LightGBM (AUC = 0.758) models, indicating the benefit of combining deep learning and radiomic features.ConclusionOur 2.5D deep learning model based on prostate MRI showed efficacy in identifying clinically significant prostate cancer, providing valuable references for clinical treatment and enhancing patient net benefit.https://www.frontiersin.org/articles/10.3389/fonc.2025.1539537/fullprostate canceraggressivenessMRIradiomicsdeep learningnomogram
spellingShingle Yalei Wang
Yuqing Xin
Baoqi Zhang
Fuqiang Pan
Xu Li
Manman Zhang
Yushan Yuan
Lei Zhang
Peiqi Ma
Bo Guan
Yang Zhang
Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
Frontiers in Oncology
prostate cancer
aggressiveness
MRI
radiomics
deep learning
nomogram
title Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
title_full Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
title_fullStr Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
title_full_unstemmed Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
title_short Assessment of prostate cancer aggressiveness through the combined analysis of prostate MRI and 2.5D deep learning models
title_sort assessment of prostate cancer aggressiveness through the combined analysis of prostate mri and 2 5d deep learning models
topic prostate cancer
aggressiveness
MRI
radiomics
deep learning
nomogram
url https://www.frontiersin.org/articles/10.3389/fonc.2025.1539537/full
work_keys_str_mv AT yaleiwang assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT yuqingxin assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT baoqizhang assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT fuqiangpan assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT xuli assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT manmanzhang assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT yushanyuan assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT leizhang assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT peiqima assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT boguan assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels
AT yangzhang assessmentofprostatecanceraggressivenessthroughthecombinedanalysisofprostatemriand25ddeeplearningmodels