Ge Jun, Sahiner Berkman, Hadjiiski Lubomir M, Chan Heang-Ping, Wei Jun, Helvie Mark A, Zhou Chuan
Department of Radiology, University of Michigan, Ann Arbor, Michigan 48109-0904, USA.
Med Phys. 2006 Aug;33(8):2975-88. doi: 10.1118/1.2211710.
We are developing a computer-aided detection (CAD) system to identify microcalcification clusters (MCCs) automatically on full field digital mammograms (FFDMs). The CAD system includes six stages: preprocessing; image enhancement; segmentation of microcalcification candidates; false positive (FP) reduction for individual microcalcifications; regional clustering; and FP reduction for clustered microcalcifications. At the stage of FP reduction for individual microcalcifications, a truncated sum-of-squares error function was used to improve the efficiency and robustness of the training of an artificial neural network in our CAD system for FFDMs. At the stage of FP reduction for clustered microcalcifications, morphological features and features derived from the artificial neural network outputs were extracted from each cluster. Stepwise linear discriminant analysis (LDA) was used to select the features. An LDA classifier was then used to differentiate clustered microcalcifications from FPs. A data set of 96 cases with 192 images was collected at the University of Michigan. This data set contained 96 MCCs, of which 28 clusters were proven by biopsy to be malignant and 68 were proven to be benign. The data set was separated into two independent data sets for training and testing of the CAD system in a cross-validation scheme. When one data set was used to train and validate the convolution neural network (CNN) in our CAD system, the other data set was used to evaluate the detection performance. With the use of a truncated error metric, the training of CNN could be accelerated and the classification performance was improved. The CNN in combination with an LDA classifier could substantially reduce FPs with a small tradeoff in sensitivity. By using the free-response receiver operating characteristic methodology, it was found that our CAD system can achieve a cluster-based sensitivity of 70, 80, and 90 % at 0.21, 0.61, and 1.49 FPs/image, respectively. For case-based performance evaluation, a sensitivity of 70, 80, and 90 % can be achieved at 0.07, 0.17, and 0.65 FPs/image, respectively. We also used a data set of 216 mammograms negative for clustered microcalcifications to further estimate the FP rate of our CAD system. The corresponding FP rates were 0.15, 0.31, and 0.86 FPs/image for cluster-based detection when negative mammograms were used for estimation of FP rates.
我们正在开发一种计算机辅助检测（CAD）系统，用于在全视野数字乳腺钼靶（FFDM）图像上自动识别微钙化簇（MCC）。该CAD系统包括六个阶段：预处理；图像增强；微钙化候选区域分割；单个微钙化的假阳性（FP）减少；区域聚类；以及聚类微钙化的FP减少。在单个微钙化的FP减少阶段，使用了截断平方和误差函数来提高我们用于FFDM的CAD系统中人工神经网络训练的效率和鲁棒性。在聚类微钙化的FP减少阶段，从每个聚类中提取形态学特征和从人工神经网络输出派生的特征。使用逐步线性判别分析（LDA）来选择特征。然后使用LDA分类器将聚类微钙化与FP区分开来。在密歇根大学收集了一个包含96例病例共192幅图像的数据集。该数据集包含96个MCC，其中28个聚类经活检证实为恶性，68个为良性。在交叉验证方案中，该数据集被分为两个独立的数据集用于CAD系统的训练和测试。当一个数据集用于训练和验证我们CAD系统中的卷积神经网络（CNN）时，另一个数据集用于评估检测性能。通过使用截断误差度量，可以加速CNN的训练并提高分类性能。CNN与LDA分类器相结合可以在灵敏度上有小的权衡的情况下大幅减少FP。通过使用自由响应接收器操作特征方法，发现我们的CAD系统在分别为0.21、0.61和1.49个FP/图像时，可以实现基于聚类的灵敏度为70%、80%和90%。对于基于病例的性能评估，在分别为0.07、0.17和0.65个FP/图像时，可以实现70%、80%和90%的灵敏度。我们还使用了一个216幅乳腺钼靶图像中无聚类微钙化的数据集来进一步估计我们CAD系统的FP率。当使用阴性乳腺钼靶图像估计FP率时，基于聚类检测的相应FP率分别为0.15、0.31和0.86个FP/图像。
