Breast density in dedicated breast computed tomography: Proposal of a classification system and interreader reliability

Radiomics in photon-counting dedicated breast CT: potential of texture analysis for breast density classification

Background

We investigated whether features derived from texture analysis (TA) can distinguish breast density (BD) in spiral photon-counting breast computed tomography (PC-BCT).

Methods

In this retrospective single-centre study, we analysed 10,000 images from 400 PC-BCT examinations of 200 patients. Images were categorised into four-level density scale (a–d) using Breast Imaging Reporting and Data System (BI-RADS)-like criteria. After manual definition of representative regions of interest, 19 texture features (TFs) were calculated to analyse the voxel grey-level distribution in the included image area. ANOVA, cluster analysis, and multinomial logistic regression statistics were used. A human readout then was performed on a subset of 60 images to evaluate the reliability of the proposed feature set.

Results

Of the 19 TFs, 4 first-order features and 7 second-order features showed significant correlation with BD and were selected for further analysis. Multinomial logistic regression revealed an overall accuracy of 80% for BD assessment. The majority of TFs systematically increased or decreased with BD. Skewness (rho -0.81), as a first-order feature, and grey-level nonuniformity (GLN, -0.59), as a second-order feature, showed the strongest correlation with BD, independently of other TFs. Mean skewness and GLN decreased linearly from density a to d. Run-length nonuniformity (RLN), as a second-order feature, showed moderate correlation with BD, but resulted in redundant being correlated with GLN. All other TFs showed only weak correlation with BD (range -0.49 to 0.49, p < 0.001) and were neglected.

Conclusion

TA of PC-BCT images might be a useful approach to assess BD and may serve as an observer-independent tool.

Breast Computed Tomography: Diagnostic Performance of the Maximum Intensity Projection Reformations as a Stand-Alone Method for the Detection and Characterization of Breast Findings

OBJECTIVES

This study aimed to evaluate the diagnostic performance of the maximum intensity projection (MIP) reformations of breast computed tomography (B-CT) images as a stand-alone method for the detection and characterization of breast imaging findings.

MATERIALS AND METHODS

A total of 160 women undergoing B-CT between August 2018 and December 2020 were retrospectively included; 80 patients with known breast imaging findings were matched with 80 patients without imaging findings according to age and amount of fibroglandular tissue (FGT). A total of 71 benign and 9 malignant lesions were included. Images were evaluated using 15-mm MIP in 3 planes by 2 radiologists with experience in B-CT. The presence of lesions and FGT were evaluated, using the BI-RADS classification. Interreader agreement and descriptive statistics were calculated.

RESULTS

The interreader agreement of the 2 readers for finding a lesion (benign or malignant) was 0.86 and for rating according to BI-RADS classification was 0.82. One of 9 cancers (11.1%) was missed by both readers due to dense breast tissue. BI-RADS 1 was correctly applied to 73 of 80 patients (91.3%) by reader 1 and to 74 of 80 patients (92.5%) by reader 2 without recognizable lesions. BI-RADS 2 or higher with a lesion in at least one of the breasts was correctly applied in 69 of 80 patients (86.3%) by both readers. For finding a malignant lesion, sensitivity was 88.9% (95% confidence interval [CI], 51.75%-99.72%) for both readers, and specificity was 99.3% (95% CI, 96.4%-100%) for reader 1 and 100% (95% CI, 97.20%-100.00%) for reader 2.

CONCLUSIONS

Evaluation of B-CT images using the MIP reformations may help to reduce the reading time with high diagnostic performance and confidence.

Applied Machine Learning in Spiral Breast-CT: Can We Train a Deep Convolutional Neural Network for Automatic, Standardized and Observer Independent Classification of Breast Density?

The aim of this study was to investigate the potential of a machine learning algorithm to accurately classify parenchymal density in spiral breast-CT (BCT), using a deep convolutional neural network (dCNN). In this retrospectively designed study, 634 examinations of 317 patients were included. After image selection and preparation, 5589 images from 634 different BCT examinations were sorted by a four-level density scale, ranging from A to D, using ACR BI-RADS-like criteria. Subsequently four different dCNN models (differences in optimizer and spatial resolution) were trained (70% of data), validated (20%) and tested on a “real-world” dataset (10%). Moreover, dCNN accuracy was compared to a human readout. The overall performance of the model with lowest resolution of input data was highest, reaching an accuracy on the “real-world” dataset of 85.8%. The intra-class correlation of the dCNN and the two readers was almost perfect (0.92) and kappa values between both readers and the dCNN were substantial (0.71–0.76). Moreover, the diagnostic performance between the readers and the dCNN showed very good correspondence with an AUC of 0.89. Artificial Intelligence in the form of a dCNN can be used for standardized, observer-independent and reliable classification of parenchymal density in a BCT examination.