Scholarlink

<p>The Prostate Imaging Quality (PI-QUAL) V2 is a simplified update of PI-QUAL V1 for assessing image quality in multiparametric prostate MRI (mpMRI), aimed at improving clinical use and reproducibility. This study evaluated inter-reader agreement of PI-QUAL V2 among basic prostate readers using multicenter mpMRI data and compared it to PI-QUAL V1. Five readers assessed T2-weighted, DWI (including ADC maps), and DCE images from 274 men (median age: 71 years, IQR: 60–78) across five centers. Inter-reader agreement was measured using weighted Gwet’s AC1 coefficients and compared using bootstrapped confidence intervals. For PI-QUAL V1, coefficients ranged from 0.659 to 0.855, indicating substantial-to-almost perfect agreement. For PI-QUAL V2, coefficients ranged from 0.821 to 0.897, showing almost perfect agreement. All 10 pairwise comparisons achieved almost perfect agreement for PI-QUAL V2, compared to only 3 of 10 for V1. Additionally, three comparisons showed non-overlapping confidence intervals, indicating statistically significant improvements with V2. PI-QUAL V2 demonstrated consistently higher inter-reader agreement than V1, with basic prostate readers achieving almost perfect agreement across multicenter mpMRI data. These findings highlight the improved reliability and clinical applicability of PI-QUAL V2.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Multicenter analysis for inter reader agreement of PIQUAL score v2 for basic readers in prostate MRI

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月2日

查看全文
  • Mujgan Orman,
  • Sabri Sirolu,
  • Mustafa Ege Seker,
  • Deniz Alis,
  • Mustafa Said Kartal,
  • Yeliz Basar,
  • Nurper Denizoglu,
  • Serpil Kurtcan,
  • Aydan Arslan,
  • Ilkay Oksuz,
  • Ercan Karaarslan

摘要

The Prostate Imaging Quality (PI-QUAL) V2 is a simplified update of PI-QUAL V1 for assessing image quality in multiparametric prostate MRI (mpMRI), aimed at improving clinical use and reproducibility. This study evaluated inter-reader agreement of PI-QUAL V2 among basic prostate readers using multicenter mpMRI data and compared it to PI-QUAL V1. Five readers assessed T2-weighted, DWI (including ADC maps), and DCE images from 274 men (median age: 71 years, IQR: 60–78) across five centers. Inter-reader agreement was measured using weighted Gwet’s AC1 coefficients and compared using bootstrapped confidence intervals. For PI-QUAL V1, coefficients ranged from 0.659 to 0.855, indicating substantial-to-almost perfect agreement. For PI-QUAL V2, coefficients ranged from 0.821 to 0.897, showing almost perfect agreement. All 10 pairwise comparisons achieved almost perfect agreement for PI-QUAL V2, compared to only 3 of 10 for V1. Additionally, three comparisons showed non-overlapping confidence intervals, indicating statistically significant improvements with V2. PI-QUAL V2 demonstrated consistently higher inter-reader agreement than V1, with basic prostate readers achieving almost perfect agreement across multicenter mpMRI data. These findings highlight the improved reliability and clinical applicability of PI-QUAL V2.