Verboom Sarah D, Kroes Jaap, Pires Santiago, Broeders Mireille J M, Sechopoulos Ioannis
Radboud University Medical Center, Department of Medical Imaging, Geert Grooteplein Zuid 10, Nijmegen, 6525 GA, the Netherlands.
ScreenPoint Medical BV, Nijmegen, the Netherlands.
Radiology. 2025 Aug;316(2):e242594. doi: 10.1148/radiol.242594.
Background Quantifying the uncertainty in artificial intelligence (AI)-based interpretations of mammograms could help AI integration in screening. Purpose To evaluate the reduction in radiologists' reading workload in mammographic screening while maintaining performance by incorporating an AI model that includes uncertainty quantification. Materials and Methods An AI model was introduced that outputs a probability of malignancy (PoM) and a measure of its uncertainty. Eight candidate uncertainty metrics, based on one or all suspicious regions, were tested. A hybrid reading approach was proposed, with recall decisions made by the model only when predictions were deemed confident; otherwise, radiologist double reading was applied. The approach was retrospectively optimized and tested with a previously unseen set of screening examinations from July 2003 to August 2018, split 50-50. Recall and cancer detection rates were compared with standard double reading. The model's area under the receiver operating characteristic curve (AUC) was compared between examinations with uncertain predictions and examinations with certain predictions. One-tailed values were obtained with bootstrapping, and the significance threshold was .05. Results The dataset included 41 469 examinations from 15 522 women; the median age was 59 years (IQR, 54.0-66.0 years). With the best-performing uncertainty metric, the entropy of the mean PoM of one region, 61.9% of the examinations were read by radiologists. Hybrid reading resulted in a recall rate of 23.6‰ (95% CI: 21.6, 25.5) and a cancer detection rate of 6.6‰ (95% CI: 5.5, 7.7), similar to that of standard double reading (23.9‰ [95% CI: 21.9, 25.8; = .27] and 6.6‰ [95% CI: 5.5, 7.7; = .14], respectively). The model's AUC was lower for examinations with uncertain predictions (0.87 [95% CI: 0.82, 0.92]) than for examinations with certain predictions (0.96 [95% CI: 0.89, 0.99]) ( = .02). Conclusion Incorporating an AI mammography interpretation model that includes uncertainty quantifications in the reading of screening mammograms may reduce the radiologists' workload substantially without changing cancer detection and recall rates. © RSNA, 2025 See also the editorial by Baltzer in this issue.
量化基于人工智能（AI）的乳房X光检查解读中的不确定性，有助于将AI整合到筛查中。目的：通过纳入一个包含不确定性量化的AI模型，评估在乳房X光筛查中减少放射科医生阅读工作量的同时保持检查性能。材料与方法：引入一个输出恶性概率（PoM）及其不确定性度量的AI模型。测试了基于一个或所有可疑区域的八个候选不确定性指标。提出了一种混合阅读方法，只有在预测被认为可信时才由模型做出召回决策；否则，应用放射科医生双读。该方法通过2003年7月至2018年8月一组之前未见过的筛查检查进行回顾性优化和测试，按50-50划分。将召回率和癌症检出率与标准双读进行比较。比较模型在预测不确定的检查和预测确定的检查之间的受试者操作特征曲线下面积（AUC）。通过自举法获得单尾P值，显著性阈值为0.05。结果：数据集包括来自15522名女性的41469次检查；中位年龄为59岁（IQR，54.0-66.0岁）。使用表现最佳的不确定性指标，即一个区域平均PoM的熵，61.9%的检查由放射科医生阅读。混合阅读的召回率为23.6‰（95%CI：21.6，25.5），癌症检出率为6.6‰（95%CI：5.5，7.7），与标准双读相似（分别为23.9‰[95%CI：21.9，25.8；P = 0.27]和6.6‰[95%CI：5.5，7.7；P = 0.14]）。模型在预测不确定的检查中的AUC（0.87[95%CI：0.82，0.92]）低于预测确定的检查（0.96[95%CI：0.89，0.99]）（P = 0.02）。结论：在筛查乳房X光检查的阅读中纳入一个包含不确定性量化的AI乳房X光解读模型，可能会大幅减少放射科医生的工作量，同时不改变癌症检出率和召回率。©RSNA，2025 另见本期Baltzer的社论。
