Diagnostic Value of Artificial Intelligence-Based Multiparametric MRI for Prostate Cancer in Patients With PSA Levels in the Gray Zone.

Authors

Bingqiang Mei, Liya Zhu, Yaojun Li, Xinsheng Mao, Guoqiu Dong

Published in

Archivos espanoles de urologia. Volume 79. Issue 8. Pages 1363-1372.

Abstract

Diagnosing prostate cancer in patients with prostate-specific antigen (PSA) levels in the gray zone (4-10 ng/mL) remains challenging. In recent years, artificial intelligence (AI) technology has been able to analyse the imaging features of multiparametric MRI (mpMRI). This study aimed to explore the diagnostic value of AI-based mpMRI technology for prostate cancer in patients with PSA levels in the gray zone.
This retrospective study included 213 patients aged 45-85 years who underwent prostate mpMRI between May 2020 and May 2025. A deep learning AI system was used to analyse the mpMRI images to obtain parameters of the primary prostate lesion. Using pathological results as the gold standard, the diagnostic value of mpMRI parameters in prostate cancer was analysed.
Among the 213 patients, 72 had prostate cancer (clinically significant prostate group) and 141 did not (non clinically significant prostate group). The mean age was (72.7 ± 7.4) years, and the mean body mass index was (25.7 ± 3.9) kg/m². Compared with the non-clinically significant prostate cancer (non-csPCa) group, the clinically significant prostate cancer group had significantly lower ellipsoid volume and segmented prostate volume (p < 0.05). Pearson analysis showed negative correlation between PSA level and segmented prostate volume in all patients with prostate cancer (r = -0.449, p < 0.01). PSA level, apparent diffusion coefficient (ADC), maximum lesion diameter, peripheral-zone signal intensity, and the prostate-specific antigen density (PSAD) derived from mpMRI exhibited strong predictive value for prostate cancer (p < 0.05).
The AI-based mpMRI can enhance diagnostic value for clinically significant prostate cancer and provide valuable insights into tumour invasiveness, potentially guiding more informed clinical decisions and reducing unnecessary biopsies.

PMID:
42876127
Bibliographic data and abstract were imported from PubMed on 15 Oct 2026.

Read full publication at:
Please sign in to see all details.

Stats

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.