From the Department of Radiology, University Hospital of Basel, Basel, Basel-Stadt, Switzerland.
Department of Radiology, NYU Langone Health, New York, NY.
Invest Radiol. 2021 Oct 1;56(10):605-613. doi: 10.1097/RLI.0000000000000780.
The aim of this study was to evaluate the effect of a deep learning based computer-aided diagnosis (DL-CAD) system on radiologists' interpretation accuracy and efficiency in reading biparametric prostate magnetic resonance imaging scans.
We selected 100 consecutive prostate magnetic resonance imaging cases from a publicly available data set (PROSTATEx Challenge) with and without histopathologically confirmed prostate cancer. Seven board-certified radiologists were tasked to read each case twice in 2 reading blocks (with and without the assistance of a DL-CAD), with a separation between the 2 reading sessions of at least 2 weeks. Reading tasks were to localize and classify lesions according to Prostate Imaging Reporting and Data System (PI-RADS) v2.0 and to assign a radiologist's level of suspicion score (scale from 1-5 in 0.5 increments; 1, benign; 5, malignant). Ground truth was established by consensus readings of 3 experienced radiologists. The detection performance (receiver operating characteristic curves), variability (Fleiss κ), and average reading time without DL-CAD assistance were evaluated.
The average accuracy of radiologists in terms of area under the curve in detecting clinically significant cases (PI-RADS ≥4) was 0.84 (95% confidence interval [CI], 0.79-0.89), whereas the same using DL-CAD was 0.88 (95% CI, 0.83-0.94) with an improvement of 4.4% (95% CI, 1.1%-7.7%; P = 0.010). Interreader concordance (in terms of Fleiss κ) increased from 0.22 to 0.36 (P = 0.003). Accuracy of radiologists in detecting cases with PI-RADS ≥3 was improved by 2.9% (P = 0.10). The median reading time in the unaided/aided scenario was reduced by 21% from 103 to 81 seconds (P < 0.001).
Using a DL-CAD system increased the diagnostic accuracy in detecting highly suspicious prostate lesions and reduced both the interreader variability and the reading time.
本研究旨在评估基于深度学习的计算机辅助诊断（DL-CAD）系统对放射科医生阅读双参数前列腺磁共振成像扫描结果的准确性和效率的影响。
我们从一个公开可用的数据集（PROSTATEx 挑战赛）中选择了 100 例连续的前列腺磁共振成像病例，这些病例既有经组织病理学证实的前列腺癌，也有无前列腺癌。7 名具有董事会认证的放射科医生在 2 个阅读块中（有和没有 DL-CAD 辅助）两次阅读每个病例，两次阅读之间的间隔至少为 2 周。阅读任务是根据前列腺成像报告和数据系统（PI-RADS）v2.0 定位和分类病变，并分配放射科医生的可疑程度评分（从 1-5 以 0.5 递增；1 为良性；5 为恶性）。通过 3 名经验丰富的放射科医生的共识阅读确定了ground truth。评估了检测性能（受试者工作特征曲线）、变异性（Fleiss κ）和没有 DL-CAD 辅助的平均阅读时间。
放射科医生在检测临床上有意义的病例（PI-RADS≥4）的曲线下面积平均准确率为 0.84（95%置信区间[CI]，0.79-0.89），而使用 DL-CAD 的准确率为 0.88（95%CI，0.83-0.94），提高了 4.4%（95%CI，1.1%-7.7%；P=0.010）。读者间的一致性（以 Fleiss κ表示）从 0.22 增加到 0.36（P=0.003）。放射科医生检测 PI-RADS≥3 的病例的准确率提高了 2.9%（P=0.10）。在无辅助/辅助场景下，阅读时间中位数从 103 秒减少到 81 秒，减少了 21%（P<0.001）。
使用 DL-CAD 系统提高了检测高度可疑前列腺病变的诊断准确性，并降低了读者间的变异性和阅读时间。
