Radiomics‐augmented lesion characterization on prostate multiparametric magnetic resonance imaging: first-order features and ratio metrics may improve diagnostic performance in prostate cancer
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background: Multiparametric magnetic resonance imaging (mpMRI) is increasingly recognized as an important diagnostic modality in the evaluation of prostate cancer. This analysis sought to investigate whether simple, ratio-normalized quantitative magnetic resonance imaging (MRI) biomarkers obtained from routine mpMRI of the prostate provide incremental diagnostic value beyond the Prostate Imaging Reporting and Data System version 2.1 (PI-RADS v2.1) for the detection of clinically significant prostate cancer (csPCa). Methods: In this retrospective single-center study, 129 men underwent pre-biopsy 3-Tesla mpMRI followed by MRI/ultrasound fusion-guided biopsy. Quantitative first-order imaging features representing the three principal imaging domains underlying PI-RADS were derived from T2-weighted imaging (T2WI), diffusion-weighted imaging (DWI), and dynamic contrast-enhanced (DCE) imaging and normalized to intra-patient reference tissue. The analyses were performed on a lesion-specific basis and stratified by anatomical zone. Zone-specific logistic regression models were constructed to assess the incremental diagnostic value of quantitative biomarkers beyond PI-RADS v2.1. Results: In the peripheral zone (PZ; 99 lesions), PI-RADS alone exhibited moderate discrimination for csPCa [area under the curve (AUC) =0.723]. An integrative model combining PI-RADS with prostate-specific antigen density (PSAD) and quantitative MRI-derived ratios from DWI, T2WI, and DCE significantly improved discrimination [AUC =0.901, 95% confidence interval (CI): 0.839-0.964]. After bootstrap-based optimism correction, the AUC decreased to 0.879. In the transition zone (TZ; 30 lesions), univariable analyses revealed no consistent differences in quantitative parameters between clinically significant and non-clinically significant lesions. PI-RADS alone yielded an AUC of 0.670, while ratio-based quantitative parameters did not provide meaningful incremental diagnostic value beyond PI-RADS v2.1. Conclusions: Simple, ratio-normalized quantitative MRI parameters integrated with PI-RADS and PSAD showed potential incremental value for detecting csPCa in the PZ. Conversely, these parameters demonstrated no statistically significant incremental diagnostic value beyond PI-RADS in the TZ in this limited cohort. These findings support a zone-aware, pragmatic approach to the quantitative augmentation of PI-RADS using routinely available MRI metrics.
Language: English
Subfield: Radiology, Nuclear Medicine and Imaging
Field: Medicine
Domain: Health Sciences
Open Access status: diamond
OpenAlex ID: W7223204256