Maintenance Notice

Due to necessary scheduled maintenance, the JMIR Publications website will be unavailable from Wednesday, July 01, 2020 at 8:00 PM to 10:00 PM EST. We apologize in advance for any inconvenience this may cause you.

Who will be affected?

Accepted for/Published in: Journal of Medical Internet Research

Date Submitted: Oct 6, 2025
Date Accepted: May 5, 2026

The final, peer-reviewed published version of this preprint can be found here:

Magnetic Resonance Imaging–Based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: Systematic Review and Meta-Analysis

Jin Y, Yuan T, Chen Z

Magnetic Resonance Imaging–Based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: Systematic Review and Meta-Analysis

J Med Internet Res 2026;28:e85360

DOI: 10.2196/85360

PMID: 42412916

MRI-based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: A Systematic Review and Meta-Analysis

  • Yanjun Jin; 
  • Tianzuo Yuan; 
  • Zhiyuan Chen

ABSTRACT

Background:

Emerged as a promising tool in PCa diagnosis. However, current studies often lack multicenter external validation, have limited sample sizes, present significant inter‑model variability, and face overfitting concerns.

Objective:

This study aims to comprehensively evaluate the diagnostic performance of magnetic resonance imaging (MRI)-based artificial intelligence (AI) models in predicting biochemical recurrence (BCR) of prostate cancer (PCa).

Methods:

Systematic searches were conducted in the PubMed, Embase, and Web of Science databases up to March 18, 2025. Studies were included that involved participants diagnosed with PCa, utilized MRI-based AI for predicting BCR, and had clearly defined reference standards. The quality of the included studies was assessed using the PROBAST+AI tool. A bivariate random-effects model was employed to pool sensitivity, specificity, and area under the curve (AUC) statistics.

Results:

A total of 23 studies were included, with 1,599 patients in internal validation and 229 patients in external validation. For pooled internal validation, sensitivity was 0.83 (95% CI: 0.78-0.88), specificity was 0.84 (95% CI: 0.77-0.89), and AUC was 0.89 (95% CI: 0.86-0.92). In external validation, sensitivity was 0.74 (95% CI: 0.63-0.83), specificity was 0.79 (95% CI: 0.58-0.91), and AUC was 0.79 (95% CI: 0.75-0.82), with a statistically significant difference in AUC (P < 0.001). MN&XGB algorithm achieved the highest sensitivity, MKL&SVM obtained the highest specificity, and the MN&XGB algorithm appeared to have the highest AUC.

Conclusions:

MRI-based AI models show potential in predicting BCR in PCa, but their performance diminishes in external validation cohorts. Patients treated with EBRT alone exhibit the highest AUC values, while those receiving EBRT with or without hormone therapy demonstrate the lowest AUC. To improve the generalizability of these models, future research should prioritize large-scale, prospective multicenter studies. Clinical Trial: PROSPERO (CRD420251102879)


 Citation

Please cite as:

Jin Y, Yuan T, Chen Z

Magnetic Resonance Imaging–Based Artificial Intelligence in Predicting Prostate Cancer Biochemical Recurrence: Systematic Review and Meta-Analysis

J Med Internet Res 2026;28:e85360

DOI: 10.2196/85360

PMID: 42412916

Download PDF


Request queued. Please wait while the file is being generated. It may take some time.

© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.