From the Institute of Diagnostic and Interventional Radiology, University Hospital Zurich, Zurich, Switzerland.
Invest Radiol. 2021 Apr 1;56(4):224-231. doi: 10.1097/RLI.0000000000000729.
Over 56,000 images of 268 mammograms from 94 patients were labeled to 3 classes according to the BI-RADS standard: "no microcalcifications" (BI-RADS 1), "probably benign microcalcifications" (BI-RADS 2/3), and "suspicious microcalcifications" (BI-RADS 4/5). Using the preprocessed images, a dCNN was trained and validated, generating 3 types of models: BI-RADS 4 cohort, BI-RADS 5 cohort, and BI-RADS 4 + 5 cohort. For the final validation of the trained dCNN models, a test data set consisting of 141 images of 51 mammograms from 26 patients labeled according to the corresponding BI-RADS classification from the radiological reports was applied. The performances of the dCNN models were evaluated, classifying each of the mammograms and computing the accuracy in comparison to the classification from the radiological reports. For visualization, probability maps of the classification were generated.
The accuracy on the validation set after 130 epochs was 99.5% for the BI-RADS 4 cohort, 99.6% for the BI-RADS 5 cohort, and 98.1% for the BI-RADS 4 + 5 cohort. Confusion matrices of the "real-world" test data set for the 3 cohorts were generated where the radiological reports served as ground truth. The resulting accuracy was 39.0% for the BI-RADS 4 cohort, 80.9% for BI-RADS 5 cohort, and 76.6% for BI-RADS 4 + 5 cohort. The probability maps exhibited excellent image quality with correct classification of microcalcification distribution.
The dCNNs can be trained to successfully classify microcalcifications on mammograms according to the BI-RADS classification system in order to act as a standardized quality control tool providing the expertise of a team of radiologists.
对 94 名患者的 268 张乳腺 X 光片的超过 56000 张图像进行了标记，根据 BI-RADS 标准分为 3 类：“无微钙化”（BI-RADS 1）、“可能良性微钙化”（BI-RADS 2/3）和“可疑微钙化”（BI-RADS 4/5）。使用预处理的图像，训练和验证了一个 dCNN，生成了 3 种类型的模型：BI-RADS 4 队列、BI-RADS 5 队列和 BI-RADS 4+5 队列。为了对训练好的 dCNN 模型进行最终验证，应用了一个由 26 名患者的 51 张乳腺 X 光片的 141 张图像组成的测试数据集，这些图像根据放射报告中的相应 BI-RADS 分类进行了标记。评估了 dCNN 模型的性能，对每张乳腺 X 光片进行分类，并与放射报告的分类进行比较，计算准确率。为了可视化，生成了分类的概率图。
经过 130 个周期的验证集准确率为 BI-RADS 4 队列 99.5%，BI-RADS 5 队列 99.6%，BI-RADS 4+5 队列 98.1%。为 3 个队列生成了“真实世界”测试数据集的混淆矩阵，放射报告作为地面实况。BI-RADS 4 队列的准确率为 39.0%，BI-RADS 5 队列为 80.9%，BI-RADS 4+5 队列为 76.6%。概率图表现出极好的图像质量，正确分类了微钙化分布。
dCNN 可以根据 BI-RADS 分类系统对乳腺 X 光片中的微钙化进行分类，从而成为一种标准化的质量控制工具，提供一组放射科医生的专业知识。
