Optimal reconstruction and quantitative image features for computer‐aided diagnosis tools for breast CT

Abstract

Purpose The purpose of this study is to determine the optimal representative reconstruction and quantitative image feature set for a computer-aided diagnosis (CADx) scheme for dedicated breast computer tomography (bCT). Method We used 93 bCT scans that contain 102 breast lesions (62 malignant, 40 benign). Using an iterative image reconstruction (IIR) algorithm, we created 37 reconstructions with different image appearances for each case. In addition, we added a clinical reconstruction for comparison purposes. We used image sharpness, determined by the gradient of gray value in a parenchymal portion of the reconstructed breast, as a surrogate measure of the image qualities/appearances for the 38 reconstructions. After segmentation of the breast lesion, we extracted 23 quantitative image features. Using leave-one-out cross-validation (LOOCV), we conducted the feature selection, classifier training, and testing. For this study, we used the linear discriminant analysis classifier. Then, we selected the representative reconstruction and feature set for the classifier with the best diagnostic performance among all reconstructions and feature sets. Then, we conducted an observer study with six radiologists using a subset of breast lesions (N = 50). Using 1000 bootstrap samples, we compared the diagnostic performance of the trained classifier to those of the radiologists. Result The diagnostic performance of the trained classifier increased as the image sharpness of a given reconstruction increased. Among combinations of reconstructions and quantitative image feature sets, we selected one of the sharp reconstructions and three quantitative image feature sets with the first three highest diagnostic performances under LOOCV as the representative reconstruction and feature set for the classifier. The classifier on the representative reconstruction and feature set achieved better diagnostic performance with an area under the ROC curve (AUC) of 0.94 (95% CI = [0.81, 0.98]) than those of the radiologists, where their maximum AUC was 0.78 (95% CI = [0.63, 0.90]). Moreover, the partial AUC, at 90% sensitivity or higher, of the classifier (pAUC = 0.085 with 95% CI = [0.063, 0.094]) was statistically better (p-value < 0.0001) than those of the radiologists (maximum pAUC = 0.009 with 95% CI = [0.003, 0.024]). Conclusion We found that image sharpness measure can be a good candidate to estimate the diagnostic performance of a given CADx algorithm. In addition, we found that there exists a reconstruction (i.e., sharp reconstruction) and a feature set that maximizes the diagnostic performance of a CADx algorithm. On this optimal representative reconstruction and feature set, the CADx algorithm outperformed radiologists.

Neutrosophic segmentation of breast lesions for dedicated breast computed tomography

“…This makes sense as weak features cannot build a classifier that can be generalizable for unseen data. This supports the finding of our previous study, 28 where we showed that a classifier trained with a set of a few strong features can achieve better classification performance than classifiers trained with a set of weak features.…”

Relationship between computer segmentation performance and computer classification performance in breast CT: A simulation study using RGI segmentation and LDA classification

“…Under an institutional review board (IRB) approved protocol, the prototype dedicated breast CT system at the University of California at Davis was used to acquire all breast CT images of recruited women. We describe the details of the cases elsewhere …”

“…Computer classification requires a choice of a statistical classifier and input features. We considered a total of 23 quantitative image features from the segmentation results (Table , adopted from). Previous studies utilized these features for lesion detection and classification …”

“…Previous studies utilized these features for lesion detection and classification. 5,10,[14][15][16][17][18][19] The 23 quantitative image features included four histogram (feature #1-#4), seven shape (feature #5-#11), five margin (feature #12-#16), four texture (feature #17-#20), and three surface curvature descriptors (feature #21-#23). Histogram descriptors characterize the gray-scale information within the lesion and its relationship to the surrounding background.…”

Optimization of computer aided detection systems: An evolutionary approach

“…Litjens et al (2017) have performed a comprehensive-review of applications of deep learning to medical image analysis; their findings confirm that most researchers employ an intuition-based random search to optimize hyper-parameters. Finally, a few authors have also sought to optimize acquisition or reconstruction parameters in order to optimize CAD performance (Lau, 2011;Lee et al, 2017). For instance, Lau (2011) developed realistic phantoms to optimize image acquisition parameters and therefore, indirectly, CAD results.In this work, we have assumed the imaging system and reconstruction algorithm to be fixed, which is the most realistic scenario for CAD development both in industry and academia wherever images are acquired by a commercial scanner in a clinical setting.…”