Search

PurposeTo develop and evaluate the diagnostic performance of an algorithm for multi-marker radiomic-based classification of breast masses in dedicated breast computed tomography (bCT) images.MethodsOver 1000 radiomic descriptors aimed at quantifying mass and border heterogeneity, morphology, and margin sharpness were developed and implemented. These included well-established texture and shape feature descriptors, which were supplemented with additional approaches for contour irregularity quantification, spicule and lobe detection, characterization of degree of infiltration, and differences in peritumoral compartments. All descriptors were extracted from a training set of 202 bCT masses (133 benign and 69 malignant), and their individual diagnostic performance was investigated in terms of area under the receiver operating characteristics (ROC) curve (AUC) of single-feature-based linear discriminant analysis (LDA) classifiers. Subsequently, the most relevant descriptors were selected through a multiple-step feature selection process (including stability analysis, statistical significance, evaluation of feature interaction, and dimensionality reduction), and used to develop a final LDA radiomic model for classification of benign and malignant masses, which was then tested on an independent test set of 82 cases (45 benign and 37 malignant).ResultsThe majority of the individual radiomic descriptors showed, on the training set, an AUC value deriving from a linear decision boundary higher than 0.65, with the lower limit of the associated 95% confidence interval (C.I.) not overlapping with random chance (AUC = 0.5). The final LDA radiomic model resulted in a test set AUC of 0.90 (95% C.I. 0.80-0.96).ConclusionsThe proposed multi-marker radiomic approach achieved high diagnostic accuracy in bCT mass classification, using a radiomic signature based on different feature types. While future studies with larger datasets are needed to further validate these results, quantitative radiomics applied to bCT shows potential to improve the breast cancer diagnosis pipeline.

PurposeClinical use of dedicated breast computed tomography (bCT) requires relatively short scan times necessitating systems with high frame rates. This in turn impacts the x-ray tube operating range. We characterize the effects of tube voltage, beam filtration, dose, and object size on contrast and noise properties related to soft tissue and iodine contrast agents as a way to optimize imaging protocols for soft tissue and iodine contrast at high frame rates.MethodsThis study design uses the signal-difference-to-noise ratio (SDNR), noise-equivalent quanta (NEQ), and detectability (d´) as measures of imaging performance for a prototype breast CT scanner that utilizes a pulsed x-ray tube (with a 4 ms pulse width) at 43.5 fps acquisition rate. We assess a range of kV, filtration, breast phantom size, and mean glandular dose (MGD). Performance measures are estimated from images of adipose-equivalent breast phantoms machined to have a representative size and shape of small, medium, and large breasts. Water (glandular tissue equivalent) and iodine contrast (5 mg/ml) were used to fill two cylindrical wells in the phantoms.ResultsAir kerma levels required for obtaining an MGD of 6 mGy ranged from 7.1 to 9.1 mGy and are reported across all kV, filtration, and breast phantom sizes. However, at 50 kV, the thick filters (0.3 mm of Cu or Gd) exceeded the maximum available mA of the x-ray generator, and hence, these conditions were excluded from subsequent analysis. There was a strong positive association between measurements of SDNR and d' (R2  > 0.97) within the range of parameters investigated in this work. A significant decrease in soft tissue SDNR was observed for increasing phantom size and increasing kV with a maximum SDNR at 50 kV with 0.2 mm Cu or 0.2 mm Gd filtration. For iodine contrast SDNR, a significant decrease was observed with increasing phantom size, but a decrease in SDNR for increasing kV was only observed for 70 kV (50 and 60 kV were not significantly different). Thicker Gd filtration (0.3 mm Gd) resulted in a significant increase in iodine SDNR and decrease in soft tissue SDNR but requires significantly more tube current to deliver the same MGD.ConclusionsThe choice of 60 kV with 0.2 mm Gd filtration provides a good trade-off for maximizing both soft tissue and iodine contrast. This scanning technique takes advantage of the ~50 keV Gd k-edge to produce contrast and can be achieved within operating range of the x-ray generator used in this work. Imaging at 60 kV allows for a greater range in dose delivered to the large breast sizes when uniform image quality is desired across all breast sizes. While imaging performance metrics (i.e., detectability index and SDNR) were shown to be strongly correlated, the methodologies presented in this work for the estimation of NEQ (and subsequently d') provides a meaningful description of the spatial resolution and noise characteristics of this prototype bCT system across a range of beam quality, dose, and object sizes.

Many computer aided diagnosis (CADx) tools for breast cancer begin by fully or semiautomatically segmenting a given breast lesion and then classifying the lesion's likelihood of malignancy using quantitative features extracted from the image. It is often assumed that better segmentation will result in better classification. However, this has not been thoroughly evaluated. The purpose of this study is to evaluate the relationship between computer segmentation performance and computer classification performance. We used 85 breast lesions (32 benign, 56 malignant) from breast computed tomography (CT) cases of 82 women. We prepared one smooth and one sharp iterative image reconstructions (IIR) and a clinical reconstruction for each of the 82 breast CT scans. For each reconstruction, we created 15 segmentation outcomes by applying 15 different segmentation algorithms. Specifically, we simulated 15 segmentation algorithms by changing parameters in a single segmentation algorithm. We then created 15 classification outcomes by conducting quantitative image feature analysis on the segmented image results. Using a 10-fold cross-validation, we evaluated the relationship between segmentation and classification performances. We found a low positive correlation between segmentation and classification performances for the smooth IIR (median Pearson's rho = 0.18), while a moderate positive correlation (median Pearson's rho = 0.4-0.43) was found between the two performances for the sharp IIR and clinical reconstruction. However, we found large variations in both segmentation and classification performances for the sharp IIR and clinical reconstruction. There were cases where segmentation algorithms resulted in similar segmentation performances, but the corresponding classification performances were different. These results indicate that an improvement in segmentation performance does not guarantee an improvement in the corresponding classification performance. Computer segmentation is an indirect variable affecting the computer classification. As better segmentation does not guarantee better classification, we should report both segmentation and classification performances when comparing segmentation algorithms.

We proposed the neutrosophic approach for segmenting breast lesions in breast computed tomography (bCT) images. The neutrosophic set considers the nature and properties of neutrality (or indeterminacy). We considered the image noise as an indeterminate component while treating the breast lesion and other breast areas as true and false components. We iteratively smoothed and contrast-enhanced the image to reduce the noise level of the true set. We then applied one existing algorithm for bCT images, the RGI segmentation, on the resulting noise-reduced image to segment the breast lesions. We compared the segmentation performance of the proposed method (named as NS-RGI) to that of the regular RGI segmentation. We used 122 breast lesions (44 benign and 78 malignant) of 111 noncontrast enhanced bCT cases. We measured the segmentation performances of the NS-RGI and the RGI using the Dice coefficient. The average Dice values of the NS-RGI and RGI were 0.82 and 0.80, respectively, and their difference was statistically significant ([Formula: see text]). We conducted a subsequent feature analysis on the resulting segmentations. The classifier performance for the NS-RGI ([Formula: see text]) improved over that of the RGI ([Formula: see text], [Formula: see text]).

The aim of this work was the evaluation of the software BreastSimulator, a breast x-ray imaging simulation software, as a tool for the creation of 3D uncompressed breast digital models and for the simulation and the optimization of computed tomography (CT) scanners dedicated to the breast. Eight 3D digital breast phantoms were created with glandular fractions in the range 10%-35%. The models are characterised by different sizes and modelled realistic anatomical features. X-ray CT projections were simulated for a dedicated cone-beam CT scanner and reconstructed with the FDK algorithm. X-ray projection images were simulated for 5 mono-energetic (27, 32, 35, 43 and 51 keV) and 3 poly-energetic x-ray spectra typically employed in current CT scanners dedicated to the breast (49, 60, or 80 kVp). Clinical CT images acquired from two different clinical breast CT scanners were used for comparison purposes. The quantitative evaluation included calculation of the power-law exponent, β, from simulated and real breast tomograms, based on the power spectrum fitted with a function of the spatial frequency, f, of the form S(f)  =  α/f   β . The breast models were validated by comparison against clinical breast CT and published data. We found that the calculated β coefficients were close to that of clinical CT data from a dedicated breast CT scanner and reported data in the literature. In evaluating the software package BreastSimulator to generate breast models suitable for use with breast CT imaging, we found that the breast phantoms produced with the software tool can reproduce the anatomical structure of real breasts, as evaluated by calculating the β exponent from the power spectral analysis of simulated images. As such, this research tool might contribute considerably to the further development, testing and optimisation of breast CT imaging techniques.

Conventional mammographic dosimetry has been developed over the past 40 years. Prior to the availability of high-resolution three-dimensional breast images, certain assumptions about breast anatomy were required. These assumptions were based on the information evident on two-dimensional mammograms; they included assumptions of thick skin, a uniform mixture of glandular and adipose tissue, and a median breast density of 50%. Recently, the availability of high-resolution breast CT studies has provided more accurate data about breast anatomy, and this, in turn, has provided the opportunity to update mammographic dosimetry. Based on hundreds of data sets on breast CT volume, a number of studies were performed and reported which have shed light on the basic breast anatomy specific to dosimetry in mammography. It was shown that the average skin thickness of the breast was approximately 1.5 mm, instead of the 4 or 5 mm in the past. In another study, 3-D breast CT data sets were used for validation of the 2-D algorithm developed at the University of Toronto, leading to data suggesting that the overall average breast density is of the order of 16-20%, rather than the previously assumed 50%. Both of these assumptions led to normalized glandular dose (DgN) coefficients which are higher than those of the past. However, a comprehensive study on hundreds of breast CT data sets confirmed the findings of other investigators that there is a more centralized average location of glandular tissue within the breast. Combined with Monte Carlo studies for dosimetry, when accurate models of the distribution of glandular tissue were used, a 30% reduction in the radiation dose (as determined by the DgN coefficient) was found as an average across typical molybdenum and tungsten spectra used clinically. The 30% average reduction was found even when the thinner skin and the lower average breast density were considered. The article reviews three specific anatomic observations made possible based on high-resolution breast CT data by several different research groups. It is noted that, periodically, previous assumptions pertaining to dosimetry can be updated when new information becomes available, so that more accurate dosimetry is achieved. Dogmatic practices typically change slowly, but it is hoped that the medical physics community will continue to evaluate changes in DgN coefficients such that they become more accurate.

PurposeThe 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).MethodWe 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.ResultThe 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

PurposeCompare conspicuity of ductal carcinoma in-situ (DCIS) to benign calcifications on unenhanced (bCT), contrast-enhanced dedicated breast CT (CEbCT) and mammography (DM).Methods and materialsThe institutional review board approved this HIPAA-compliant study. 42 women with Breast Imaging Reporting and Data System 4 or 5 category micro-calcifications had breast CT before biopsy. Three subjects with invasive disease at surgery were excluded. Two breast radiologists independently compared lesion conspicuity scores (CS) for CEbCT, to bCT and DM. Enhancement was measured in Hounsfield units (HU). Mean CS ± standard deviations are shown. Receiver operating characteristic analysis (ROC) measured radiologists' discrimination performance by comparing CS to enhancement alone. Statistical measurements were made using ANOVA F-test, Wilcoxon rank-sum test and robust linear regression analyses.Results39 lesions (17 DCIS, 22 benign) were analyzed. DCIS (8.5 ± 0.9, n=17) was more conspicuous than benign micro-calcifications (3.6 ± 2.9, n=22; p

PurposeCurrent dosimetry methods in mammography assume that the breast is comprised of a homogeneous mixture of glandular and adipose tissues. Three-dimensional (3D) dedicated breast CT (bCT) data sets were used previously to assess the complex anatomical structure within the breast, characterizing the statistical distribution of glandular tissue in the breast. The purpose of this work was to investigate the effect of bCT-derived heterogeneous glandular distributions on dosimetry in mammography.MethodsbCT-derived breast diameters, volumes, and 3D fibroglandular distributions were used to design realistic compressed breast models comprised of heterogeneous distributions of glandular tissue. The bCT-derived glandular distributions were fit to biGaussian functions and used as probability density maps to assign the density distributions within compressed breast models. The MCNPX 2.6.0 Monte Carlo code was used to estimate monoenergetic normalized mean glandular dose "DgN(E)" values in mammography geometry. The DgN(E) values were then weighted by typical mammography x-ray spectra to determine polyenergetic DgN (pDgN) coefficients for heterogeneous (pDgNhetero) and homogeneous (pDgNhomo) cases. The dependence of estimated pDgN values on phantom size, volumetric glandular fraction (VGF), x-ray technique factors, and location of the heterogeneous glandular distributions was investigated.ResultsThe pDgNhetero coefficients were on average 35.3% (SD, 4.1) and 24.2% (SD, 3.0) lower than the pDgNhomo coefficients for the Mo-Mo and W-Rh x-ray spectra, respectively, across all phantom sizes and VGFs when the glandular distributions were centered within the breast phantom in the coronal plane. At constant breast size, increasing VGF from 7.3% to 19.1% lead to a reduction in pDgNhetero relative to pDgNhomo of 23.6%-27.4% for a W-Rh spectrum. Displacement of the glandular distribution, at a distance equal to 10% of the compressed breast width in the superior and inferior directions, resulted in a 37.3% and a -26.6% change in the pDgNhetero coefficient, respectively, relative to the centered distribution for the Mo-Mo spectrum. Lateral displacement of the glandular distribution, at a distance equal to 10% of the compressed breast width, resulted in a 1.5% change in the pDgNhetero coefficient relative to the centered distribution for the W-Rh spectrum.ConclusionsIntroducing bCT-derived heterogeneous glandular distributions into mammography phantom design resulted in decreased glandular dose relative to the widely used homogeneous assumption. A homogeneous distribution overestimates the amount of glandular tissue near the entrant surface of the breast, where dose deposition is exponentially higher. While these findings are based on clinically measured distributions of glandular tissue using a large cohort of women, future work is required to improve the classification of glandular distributions based on breast size and overall glandular fraction.

PURPOSE:The purpose of this study is to measure the effectiveness of local curvature measures as novel image features for classifying breast tumors. METHODS:A total of 119 breast lesions from 104 noncontrast dedicated breast computed tomography images of women were used in this study. Volumetric segmentation was done using a seed-based segmentation algorithm and then a triangulated surface was extracted from the resulting segmentation. Total, mean, and Gaussian curvatures were then computed. Normalized curvatures were used as classification features. In addition, traditional image features were also extracted and a forward feature selection scheme was used to select the optimal feature set. Logistic regression was used as a classifier and leave-one-out cross-validation was utilized to evaluate the classification performances of the features. The area under the receiver operating characteristic curve (AUC, area under curve) was used as a figure of merit. RESULTS:Among curvature measures, the normalized total curvature (CT) showed the best classification performance (AUC of 0.74), while the others showed no classification power individually. Five traditional image features (two shape, two margin, and one texture descriptors) were selected via the feature selection scheme and its resulting classifier achieved an AUC of 0.83. Among those five features, the radial gradient index (RGI), which is a margin descriptor, showed the best classification performance (AUC of 0.73). A classifier combining RGI and CT yielded an AUC of 0.81, which showed similar performance (i.e., no statistically significant difference) to the classifier with the above five traditional image features. Additional comparisons in AUC values between classifiers using different combinations of traditional image features and CT were conducted. The results showed that CT was able to replace the other four image features for the classification task. CONCLUSIONS:The normalized curvature measure contains useful information in classifying breast tumors. Using this, one can reduce the number of features in a classifier, which may result in more robust classifiers for different datasets.