Characterization of masses in digital breast tomosynthesis: comparison of machine learning in projection views and reconstructed slices.

PURPOSE: In digital breast tomosynthesis (DBT), quasi-three-dimensional (3D) structural information is reconstructed from a small number of 2D projection view (PV) mammograms acquired over a limited angular range. The authors developed preliminary computer-aided diagnosis (CADx) methods for classification of malignant and benign masses and compared the effectiveness of analyzing lesion characteristics in the reconstructed DBT slices and in the PVs. METHODS: A data set of MLO view DBT of 99 patients containing 107 masses (56 malignant and 51 benign) was collected at the Massachusetts General Hospital with IRB approval. The DBTs were obtained with a GE prototype system which acquired 11 PVs over a 50 degree arc. The authors reconstructed the DBTs at 1 mm slice interval using a simultaneous algebraic reconstruction technique. The region of interest (ROI) containing the mass was marked by a radiologist in the DBT volume and the corresponding ROIs on the PVs were derived based on the imaging geometry. The subsequent processes were fully automated. For classification of masses using the DBT-slice approach, the mass on each slice was segmented by an active contour model initialized with adaptive k-means clustering. A spiculation likelihood map was generated by analysis of the gradient directions around the mass margin and spiculation features were extracted from the map. The rubber band straightening transform (RBST) was applied to a band of pixels around the segmented mass boundary. The RBST image was enhanced by Sobel filtering in the horizontal and vertical directions, from which run-length statistics texture features were extracted. Morphological features including those from the normalized radial length were designed to describe the mass shape. A feature space composed of the spiculation features, texture features, and morphological features extracted from the central slice alone and seven feature spaces obtained by averaging the corresponding features from three to 19 slices centered at the central slice were compared. For classification of masses using the PV approach, a feature extraction process similar to that described above for the DBT approach was performed on the ROIs from the individual PVs. Six feature spaces obtained from the central PV alone and by averaging the corresponding features from three to 11 PVs were formed. In each feature space for either the DBT-slice or the PV approach, a linear discriminant analysis classifier with stepwise feature selection was trained and tested using a two-loop leave-one-case-out resampling procedure. Simplex optimization was used to guide feature selection automatically within the training set in each leave-one-case-out cycle. The performance of the classifiers was evaluated by the area (Az) under the receiver operating characteristic curve. RESULTS: The test Az values from the DBT-slice approach ranged from 0.87 +/- 0.03 to 0.93 +/- 0.02, while those from the PV approach ranged from 0.78 +/- 0.04 to 0.84 +/- 0.04. The highest test Az of 0.93 +/- 0.02 from the nine-DBT-slice feature space was significantly (p = 0.006) better than the highest test Az of 0.84 +/- 0.04 from the nine-PV feature space. CONCLUSION: The features of breast lesions extracted from the DBT slices consistently provided higher classification accuracy than those extracted from the PV images.

Computer-aided diagnosis of lung nodules on CT scans: ROC study of its effect on radiologists' performance.

RATIONALE AND OBJECTIVES: The aim of this study was to evaluate the effect of computer-aided diagnosis (CAD) on radiologists' estimates of the likelihood of malignancy of lung nodules on computed tomographic (CT) imaging. METHODS AND MATERIALS: A total of 256 lung nodules (124 malignant, 132 benign) were retrospectively collected from the thoracic CT scans of 152 patients. An automated CAD system was developed to characterize and provide malignancy ratings for lung nodules on CT volumetric images. An observer study was conducted using receiver-operating characteristic analysis to evaluate the effect of CAD on radiologists' characterization of lung nodules. Six fellowship-trained thoracic radiologists served as readers. The readers rated the likelihood of malignancy on a scale of 0% to 100% and recommended appropriate action first without CAD and then with CAD. The observer ratings were analyzed using the Dorfman-Berbaum-Metz multireader, multicase method. RESULTS: The CAD system achieved a test area under the receiver-operating characteristic curve (A(z)) of 0.857 +/- 0.023 using the perimeter, two nodule radii measures, two texture features, and two gradient field features. All six radiologists obtained improved performance with CAD. The average A(z) of the radiologists improved significantly (P < .01) from 0.833 (range, 0.817-0.847) to 0.853 (range, 0.834-0.887). CONCLUSION: CAD has the potential to increase radiologists' accuracy in assessing the likelihood of malignancy of lung nodules on CT imaging.

Computer-aided diagnosis of pulmonary nodules on CT scans: improvement of classification performance with nodule surface features.

The purpose of this work is to develop a computer-aided diagnosis (CAD) system to differentiate malignant and benign lung nodules on CT scans. A fully automated system was designed to segment the nodule from its surrounding structured background in a local volume of interest (VOI) and to extract image features for classification. Image segmentation was performed with a 3D active contour method. The initial contour was obtained as the boundary of a binary object generated by k-means clustering within the VOI and smoothed by morphological opening. A data set of 256 lung nodules (124 malignant and 132 benign) from 152 patients was used in this study. In addition to morphological and texture features, the authors designed new nodule surface features to characterize the lung nodule surface smoothness and shape irregularity. The effects of two demographic features, age and gender, as adjunct to the image features were also investigated. A linear discriminant analysis (LDA) classifier built with features from stepwise feature selection was trained using simplex optimization to select the most effective features. A two-loop leave-one-out resampling scheme was developed to reduce the optimistic bias in estimating the test performance of the CAD system. The area under the receiver operating characteristic curve, A(z), for the test cases improved significantly (p < 0.05) from 0.821 +/- 0.026 to 0.857 +/- 0.023 when the newly developed image features were included with the original morphological and texture features. A similar experiment performed on the data set restricted to primary cancers and benign nodules, excluding the metastatic cancers, also resulted in an improved test A(z), though the improvement did not reach statistical significance (p = 0.07). The two demographic features did not significantly affect the performance of the CAD system (p > 0.05) when they were added to the feature space containing the morphological, texture, and new gradient field and radius features. To investigate if a support vector machine (SVM) classifier can achieve improved performance over the LDA classifier, we compared the performance of the LDA and SVMs with various kernels and parameters. Principal component analysis was used to reduce the dimensionality of the feature space for both the LDA and the SVM classifiers. When the number of selected principal components was varied, the highest test A(z) among the SVMs of various kernels and parameters was slightly higher than that of the LDA in one-loop leave-one-case-out resampling. However, no SVM with fixed architecture consistently performed better than the LDA in the range of principal components selected. This study demonstrated that the authors' proposed segmentation and feature extraction techniques are promising for classifying lung nodules on CT images.

Quantitative CT of lung nodules: dependence of calibration on patient body size, anatomic region, and calibration nodule size for single- and dual-energy techniques.

Calcium concentration may be a useful feature for distinguishing benign from malignant lung nodules in computer-aided diagnosis. The calcium concentration can be estimated from the measured CT number of the nodule and a CT number vs calcium concentration calibration line that is derived from CT scans of two or more calcium reference standards. To account for CT number nonuniformity in the reconstruction field, such calibration lines may be obtained at multiple locations within lung regions in an anthropomorphic phantom. The authors performed a study to investigate the effects of patient body size, anatomic region, and calibration nodule size on the derived calibration lines at ten lung region positions using both single energy (SE) and dual energy (DE) CT techniques. Simulated spherical lung nodules of two concentrations (50 and 100 mg/cc CaCO3) were employed. Nodules of three different diameters (4.8, 9.5, and 16 mm) were scanned in a simulated thorax section representing the middle of the chest with large lung regions. The 4.8 and 9.5 mm nodules were also scanned in a section representing the upper chest with smaller lung regions. Fat rings were added to the peripheries of the phantoms to simulate larger patients. Scans were acquired on a GE-VCT scanner at 80, 120, and 140 kVp and were repeated three times for each condition. The average absolute CT number separations between the calibration lines were computed. In addition, under- or overestimates were determined when the calibration lines for one condition (e.g., small patient) were used to estimate the CaCO3 concentrations of nodules for a different condition (e.g., large patient). The authors demonstrated that, in general, DE is a more accurate method for estimating the calcium contents of lung nodules. The DE calibration lines within the lung field were less affected by patient body size, calibration nodule size, and nodule position than the SE calibration lines. Under- or overestimates in CaCO3 concentrations of nodules were also in general smaller in quantity with DE than with SE. However, because the slopes of the calibration lines for DE were about one-half the slopes for SE, the relative improvement in the concentration estimates for DE as compared to SE was about one-half the relative improvement in the separation between the calibration lines. Results in the middle of the chest thorax section with large lungs were nearly completely consistent with the above generalization. On the other hand, results in the upper-chest thorax section with smaller lungs and greater amounts of muscle and bone were mixed. A repeat of the entire study in the upper thorax section yielded similar mixed results. Most of the inconsistencies occurred for the 4.8 mm nodules and may be attributed to errors caused by beam hardening, volume averaging, and insufficient sampling. Targeted, higher resolution reconstructions of the smaller nodules, application of high atomic number filters to the high energy x-ray beam for improved spectral separation, and other future developments in DECT may alleviate these problems and further substantiate the superior accuracy of DECT in quantifying the calcium concentrations of lung nodules.

Accuracy of the CT numbers of simulated lung nodules imaged with multi-detector CT scanners.

A study was performed to determine the accuracies and reproducibilities of the CT numbers of simulated lung nodules imaged with multi-detector CT scanners. The nodules were simulated by spherical balls of three diameters (4.8, 9.5, and 16 mm) and two compositions (50 and 100 mg/cc CaCO3 in water-equivalent plastic). All were scanned in a liquid-water-filled container at the center of a water-equivalent-plastic phantom and in air cavities within the same phantom using GE multi-detector CT scanners. The nodules were also scanned within simulated lung regions in an anthropomorphic thorax section phantom that was bolused on both sides with water-equivalent slabs. Results were compared for three scanning protocols--the protocol for the National Lung Screening Trial (NLST), the protocol for the Lung Tissue Research Consortium (LTRC) study, and a high resolution (small pitch, thin slice and small scan interval) higher dose "gold standard" protocol. Scans were repeated three times with each protocol to assess reproducibility. The CT numbers of the nodules in water were found to be nearly independent of nodule size. However, the presence and the size of an air cavity surrounding a nodule had a significant effect (e.g., the CT number of a 50 mg/cc nodule was 64 HU in water, 37 HU in a 1.8 cm diameter air cavity, and 19 HU in a 4.4 cm diameter air cavity). This variability of CT number with size of air cavity may affect the results of the LTRC study in which patients are scanned at both full inspiration and full expiration. The CT numbers of the 9.5 and 16 mm diameter nodules within the anthropomorphic phantom were highly reproducible (average standard deviations of 2 HU or less) for all protocols. On the other hand, both accuracy and reproducibility were significantly degraded for the 4.8 mm diameter nodules, especially for the NLST (2.5 mm thickness, 2 mm slice interval) technique. Use of thinner slice (1.25 mm) and slice interval (1.25 mm) scans that can be reconstructed retrospectively from the multi-detector helical CT projection data of the standard NLST protocol yield CT numbers for the 4.8 mm diameter nodules that are more accurate and reproducible than those of the standard NLST technique. In general, the CT numbers of the nodules were found to be lower at positions near the centers of the lungs and near the spine, which is probably due to increased beam hardening in those regions. Also, larger nodules were found to have higher CT numbers than smaller nodules, consistent with results obtained on early single slice GE CT scanners. Until manufacturers develop quantitative CT scanners with improved x-ray beam hardening and scatter corrections, it is recommended that reference phantoms be employed to more accurately assess the calcium contents of patient lung nodules in screening and tissue characterization studies and in eventual computer-aided detection and diagnosis applications.

Computer-aided diagnosis of pulmonary nodules on CT scans: segmentation and classification using 3D active contours.

We are developing a computer-aided diagnosis (CAD) system to classify malignant and benign lung nodules found on CT scans. A fully automated system was designed to segment the nodule from its surrounding structured background in a local volume of interest (VOI) and to extract image features for classification. Image segmentation was performed with a three-dimensional (3D) active contour (AC) method. A data set of 96 lung nodules (44 malignant, 52 benign) from 58 patients was used in this study. The 3D AC model is based on two-dimensional AC with the addition of three new energy components to take advantage of 3D information: (1) 3D gradient, which guides the active contour to seek the object surface, (2) 3D curvature, which imposes a smoothness constraint in the z direction, and (3) mask energy, which penalizes contours that grow beyond the pleura or thoracic wall. The search for the best energy weights in the 3D AC model was guided by a simplex optimization method. Morphological and gray-level features were extracted from the segmented nodule. The rubber band straightening transform (RBST) was applied to the shell of voxels surrounding the nodule. Texture features based on run-length statistics were extracted from the RBST image. A linear discriminant analysis classifier with stepwise feature selection was designed using a second simplex optimization to select the most effective features. Leave-one-case-out resampling was used to train and test the CAD system. The system achieved a test area under the receiver operating characteristic curve (A(z)) of 0.83 +/- 0.04. Our preliminary results indicate that use of the 3D AC model and the 3D texture features surrounding the nodule is a promising approach to the segmentation and classification of lung nodules with CAD. The segmentation performance of the 3D AC model trained with our data set was evaluated with 23 nodules available in the Lung Image Database Consortium (LIDC). The lung nodule volumes segmented by the 3D AC model for best classification were generally larger than those outlined by the LIDC radiologists using visual judgment of nodule boundaries.