Radiomic biomarker extracted from PI-RADS 3 patients support more e\`icient and robust prostate cancer diagnosis: a multi-center study

TL;DR

This study develops radiomic biomarkers from PI-RADS 3 prostate MRI patients, demonstrating improved diagnostic performance and robustness across multiple data distributions, aiding clinical decision-making.

Contribution

Introduces novel radiomic biomarkers specifically for PI-RADS 3 patients, addressing the gap in biomarker mining for hard samples in prostate cancer diagnosis.

Findings

HS biomarkers outperform traditional features across datasets

Biomarkers enhance diagnostic accuracy for PI-RADS 3 patients

Robustness of biomarkers across different data distributions

Abstract

Prostate Imaging Reporting and Data System (PI-RADS) based on multi-parametric MRI classi\^ees patients into 5 categories (PI-RADS 1-5) for routine clinical diagnosis guidance. However, there is no consensus on whether PI-RADS 3 patients should go through biopsies. Mining features from these hard samples (HS) is meaningful for physicians to achieve accurate diagnoses. Currently, the mining of HS biomarkers is insu\`icient, and the e\'eectiveness and robustness of HS biomarkers for prostate cancer diagnosis have not been explored. In this study, biomarkers from di\'eerent data distributions are constructed. Results show that HS biomarkers can achieve better performances in di\'eerent data distributions.