Dataset Information

A transfer learning nomogram for predicting prostate cancer and benign conditions on MRI.

ABSTRACT:

Background

Deep learning has been used to detect or characterize prostate cancer (PCa) on medical images. The present study was designed to develop an integrated transfer learning nomogram (TLN) for the prediction of PCa and benign conditions (BCs) on magnetic resonance imaging (MRI).

Methods

In this retrospective study, a total of 709 patients with pathologically confirmed PCa and BCs from two institutions were included and divided into training (n = 309), internal validation (n = 200), and external validation (n = 200) cohorts. A transfer learning signature (TLS) that was pretrained with the whole slide images of PCa and fine-tuned on prebiopsy MRI images was constructed. A TLN that integrated the TLS, the Prostate Imaging-Reporting and Data System (PI-RADS) score, and the clinical factor was developed by multivariate logistic regression. The performance of the TLS, clinical model (CM), and TLN were evaluated in the validation cohorts using the receiver operating characteristic (ROC) curve, the Delong test, the integrated discrimination improvement (IDI), and decision curve analysis.

Results

TLS, PI-RADS score, and age were selected for TLN construction. The TLN yielded areas under the curve of 0.9757 (95% CI, 0.9613-0.9902), 0.9255 (95% CI, 0.8873-0.9638), and 0.8766 (95% CI, 0.8267-0.9264) in the training, internal validation, and external validation cohorts, respectively, for the discrimination of PCa and BCs. The TLN outperformed the TLS and the CM in both the internal and external validation cohorts. The decision curve showed that the TLN added more net benefit than the CM.

Conclusions

The proposed TLN has the potential to be used as a noninvasive tool for PCa and BCs differentiation.

SUBMITTER: Chen J

PROVIDER: S-EPMC10691068 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

A transfer learning nomogram for predicting prostate cancer and benign conditions on MRI.

Chen Junhao J Feng Bao B Hu Maoqing M Huang Feidong F Chen Yehang Y Ma Xilun X Long Wansheng W

BMC medical imaging 20231130 1

<h4>Background</h4>Deep learning has been used to detect or characterize prostate cancer (PCa) on medical images. The present study was designed to develop an integrated transfer learning nomogram (TLN) for the prediction of PCa and benign conditions (BCs) on magnetic resonance imaging (MRI).<h4>Methods</h4>In this retrospective study, a total of 709 patients with pathologically confirmed PCa and BCs from two institutions were included and divided into training (n = 309), internal validation (n ...[more]

PMID: 38036991

Similar Datasets

Project description:ObjectiveThe aim of this study was to establish a predictive nomogram for predicting prostate cancer (PCa) in patients with gray-zone prostate-specific antigen (PSA) levels (4-10.0 ng/mL) based on radiomics and other traditional clinical parameters.MethodsIn all, 274 patients with gray-zone PSA levels were included in this retrospective study. They were randomly divided into training and validation sets (n = 191 and 83, respectively). Data on the clinical risk factors related to PCa with gray-zone PSA levels (such as Prostate Imaging Reporting and Data System, version 2.1 [PI-RADS V2.1] category, age, prostate volume, and serum PSA level) were collected for all patients. Lesion volumes of interest (VOI) from T2-weighted imaging (T2WI) and apparent diffusion coefficient (ADC) imaging were annotated by two radiologists. The radiomics model, clinical model, and combined prediction model, which was presented on a nomogram by incorporating the radiomics signature and clinical and radiological risk factors for PCa, were developed using logistic regression. The area under the receiver operator characteristic (AUC-ROC) and decision, calibration curve were used to compare the three models for the diagnosis of PCa with gray-zone PSA levels.ResultsThe predictive nomogram (AUC: 0.953) incorporating the radiomics score and PI-RADS V2.1 category, age, and the radiomics model (AUC: 0.941) afforded much higher diagnostic efficacy than the clinical model (AUC: 0.866). The addition of the rad score could improve the discriminatory performance of the clinical model. The decision curve analysis indicated that the radiomics or combined model could be more beneficial compared to the clinical model for the prediction of PCa. The nomogram showed good agreement for detecting PCa with gray-zone PSA levels between prediction and histopathologic confirmation.ConclusionThe nomogram, which combined the radiomics score and PI-RADS V2.1 category and age, is an effective and non-invasive method for predicting PCa. Furthermore, as well as good calibration and is clinically useful, which could reduce unnecessary prostate biopsies in patients having PCa with gray-zone PSA levels.

Dataset Information

A transfer learning nomogram for predicting prostate cancer and benign conditions on MRI.

Background

Methods

Results

Conclusions

Publications

A transfer learning nomogram for predicting prostate cancer and benign conditions on MRI.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets