Does image reduction affect the diagnostic accuracy of digital mammograms?

This paper attempts to pinpoint different techniques for Pectoral Muscle (PM) segmentation, Microcalcification (MC) detection and classification in digital mammograms. The segmentation of PM and detection of MC and its classification are mostly based on image processing and data mining techniques. The review centered on major techniques in image processing and data mining that is employed for PM segmentation, MC detection and classification in digital mammograms. Breast cancer is one of the significant causes of death among women aged above 40. Mammography is considered the most successful means for prompt and timely detection of breast cancers. One notable visual indication of the malignant growth is the appearance of Masses, Architectural Distortions, and Microcalcification Clusters (MCCs). There are some disadvantages and hurdles for mankind viewers, and it is hard for radiologists to supply both precise and steady assessment for a large number of mammograms created in extensive screening. Computer Aided Detection has been employed to help radiologists in detecting MC and MCCs. The automatic recognition of malignant MCCs could be very helpful for diagnostic purpose. In this paper, we summarize the methods of automatic detection and classification of MCs in digitized mammograms. Pectoral muscle segmentation techniques are also summarized. The techniques used for segmentation of PM, MC detection and classification in a digitized mammogram are reviewed.

In this article the authors evaluate a recently proposed variable dose (VD)-digital breast tomosynthesis (DBT) acquisition technique in terms of the detection accuracy for breast masses and microcalcification (MC) clusters. With this technique, approximately half of the total dose is used for one center projection and the remaining dose is split among the other tomosynthesis projection views. This acquisition method would yield both a projection view and a reconstruction view. One of the aims of this study was to evaluate whether the center projection alone of the VD acquisition can provide equal or superior MC detection in comparison to the 3D images from uniform dose (UD)-DBT. Another aim was to compare the mass-detection capabilities of 3D reconstructions from VD-DBT and UD-DBT. In a localization receiver operating characteristic (LROC) observer study of MC detection, the authors compared the center projection of a VD acquisitioh scheme (at 2 mGy dose) with detector pixel size of 100 microm with the UD-DBT reconstruction (at 4 mGy dose) obtained with a voxel size of 100 microm. MCs with sizes of 150 and 180 microm were used in the study, with each cluster consisting of seven MCs distributed randomly within a small volume. Reconstructed images in UD-DBT were obtained from a projection set that had a total of 4 mGy dose. The current study shows that for MC detection, using the center projection alone of VD acquisition scheme performs worse with area under the LROC curve (AL) of 0.76 than when using the 3D reconstructed image using the UD acquisition scheme (AL=0.84). A 2D ANOVA found a statistically significant difference (p=0.038) at a significance level of 0.05. In the current study, although a reconstructed image was also available using the VD acquisition scheme, it was not used to assist the MC detection task which was done using the center projection alone. In the case of evaluation of detection accuracy of masses, the reconstruction with VD-DBT (AL=0.71) was compared to that obtained from the UD-DBT (AL=0.78). The authors found no statistically significant difference between the two (p-value=0.22), although all the observers performed better for UD-DBT.

Mammography is one of the most efficient imaging techniques for the early detection and diagnosis of breast cancer in woman. The appearance of microcalcifications (MCs) in X-ray mammograms can be regarded as important early diagnostic feature of breast cancer. However, the detection of MCs in X-ray mammograms only depend on radiologists by visual inspection is a difficult task. In this paper, a novel algorithm for the automated detection of MCs in digital mammograms is described. The aim of approach is to locate suspicious MCs by analyzing the distribution of brightness over the digital mammograms. The algorithm for the detection of MCs is performed in two steps. First, the digital X-ray mammogram intensity surface is well-fitted by the least squares support vector machines (LS-SVM), and then the maximum extremum points are detected on the fitted intensity surface by convolving the image with the second order directional derivative operators deduced from the mapped LS-SVM with mixture of kernels. The proposed approach is developed and evaluated by using MIAS database. The experimental results demonstrate the possible MCs are detected accurately and efficiently.

Mammography is the only technique currently used for detecting microcalcification (MC) clusters, an early indicator of breast cancer. However, mammographic images superimpose a three-dimensional compressed breast image onto two-dimensional projection views, resulting in overlapped anatomical breast structures that may obscure the detection and visualization of MCs. One possible solution to this problem is the use of cone beam computed tomography (CBCT) with a flat-panel (FP) digital detector. Although feasibility studies of CBCT techniques for breast imaging have yielded promising results, they have not shown how radiation dose and x-ray tube voltage affect the accuracy with which MCs are detected by CBCT experimentally. We therefore conducted a phantom study using a FP-based CBCT system with various mean glandular doses and kVp values. An experimental CBCT scanner was constructed with a data acquisition rate of 7.5 frames/s. 10.5 and 14.5 cm diameter breast phantoms made of gelatin were used to simulate uncompressed breasts consisting of 100% glandular tissue. Eight different MC sizes of calcium carbonate grains, ranging from 180-200 microm to 355-425 microm, were used to simulate MCs. MCs of the same size were arranged to form a 5 x 5 MC cluster and embedded in the breast phantoms. These MC clusters were positioned at 2.8 cm away from the center of the breast phantoms. The phantoms were imaged at 60, 80, and 100 kVp. With a single scan (360 degrees), 300 projection images were acquired with 0.5 x, 1x, and 2x mean glandular dose limit for 10.5 cm phantom and with 1x, 2x, and 4x for 14.5 cm phantom. A Feldkamp algorithm with a pure ramp filter was used for image reconstruction. The normalized noise level was calculated for each x-ray tube voltage and dose level. The image quality of the CBCT images was evaluated by counting the number of visible MCs for each MC cluster for various conditions. The average percentage of the visible MCs was computed and plotted as a function of the MGD, the kVp, and the average MC size. The results showed that the MC visibility increased with the MGD significantly but decreased with the breast size. The results also showed that the x-ray tube voltage affects the detection of MCs under different circumstances. With a 50% threshold, the minimum detectable MC sizes for the 10.5 cm phantom were 348(+/-2), 288(+/-7), 257(+/-2) microm at 3, 6, and 12 mGy, respectively. Those for the 14.5 cm phantom were 355 (+/-1), 307 (+/-7), 275 (+/-5) microm at 6, 12, and 24 mGy, respectively. With a 75% threshold, the minimum detectable MC sizes for the 10.5 cm phantom were 367 (+/-1), 316 (+/-7), 265 (+/-3) microm at 3, 6, and 12 mGy, respectively. Those for the 14.5 cm phantom were 377 (+/-3), 334 (+/-5), 300 (+/-2) microm at 6, 12, and 24 mGy, respectively.

This paper presents a machine learning based approach for the discrimination of malignant and benign microcalcification (MC) clusters in digital mammograms. A series of morphological operations was carried out to facilitate the feature extraction from segmented microcalcification. A combination of morphological, texture, and distribution features from individual MC components and MC clusters were extracted and a correlation-based feature selection technique was used. The clinical relevance of the selected features is discussed. The proposed method was evaluated using three different databases: Optimam Mammography Image Database (OMI-DB), Digital Database for Screening Mammography (DDSM), and Mammographic Image Analysis Society (MIAS) database. The best classification accuracy (%) was achieved for OPTIMAM using a stack generalization classifier with 10-fold cross validation obtaining an A value equal to .

Spectral Phase Based Medical Image Processing 1

This paper presents a computer-aided diagnosis (CAD) system for automatic detection of clustered microcalcifications (MCs) in digitized mammograms. The proposed system consists of two main steps. First, potential MC pixels in the mammograms are segmented out by using four mixed features consisting of two wavelet features and two gray level statistical features and then the potential MC pixels are labeled into potential individual MC objects by their spatial connectivity. Second, MCs are detected by extracting a set of 17 features from the potential individual MC objects. The classifier which is used in the first step is a multilayer feedforward neural network classifier but for the second step we have used Adaboost with SVM-based component classifier. Component classifiers which we used in our combining method are SVM classifiers with RBF kernel. The method was applied to a database of 40 mammograms (Nijmegen database) containing 105 clusters of MCs. A free-response operating characteristics (FROC) curve is used to evaluate the performance of CAD system. Results show that the proposed system gives quite satisfactory performance. In particular, 89.55% mean true positive detection rate is achieved at the cost of 0.921 false positive per image.

Microcalcification (MC) clusters in mammograms can be important early signs of breast cancer in women. Accurate detection of MC clusters is an important but challenging problem. In this paper, we propose the use of a recently developed machine learning technique -- relevance vector machine (RVM) -- for automatic detection of MCs in digitized mammograms. RVM is based on Bayesian estimation theory, and as a feature it can yield a decision function that depends on only a very small number of so-called <i>relevance vectors</i>. We formulate MC detection as a supervised-learning problem, and use RVM to classify if an MC object is present or not at each location in a mammogram image. MC clusters are then identified by grouping the detected MC objects. The proposed method is tested using a database of 141 clinical mammograms, and compared with a support vector machine (SVM) classifier which we developed previously. The detection performance is evaluated using the free-response receiver operating characteristic (FROC) curves. It is demonstrated that the RVM classifier matches closely with the SVM classifier in detection performance, and does so with a much sparser kernel representation than the SVM classifier. Consequently, the RVM classifier greatly reduces the computational complexity, making it more suitable for real-time processing of MC clusters in mammograms.

Microcalcification (MC) clusters in mammograms are one of the primary signs of breast cancer. In the literature, most MC detection methods follow a two-step paradigm: segmenting each MC and analyzing their spatial distributions to form MC clusters. However, segmentation of MCs cannot avoid low sensitivity or high false positive rate due to their variability in size (sometimes <0.1 mm), brightness, and shape (with diverse surroundings). In this paper, we propose a novel self-adversarial learning framework to differentiate and delineate the MC clusters in an end-to-end manner. The class activation mapping (CAM) mechanism is employed to directly generate the contours of MC clusters with the guidance of MC cluster classification and box annotations. We also propose the self-adversarial learning strategy to equip CAM with better detection capability of MC clusters by using the backbone network itself as a discriminator. Experimental results suggest that our method can achieve better performance for MC cluster detection with the contouring of MC clusters and classification of MC types.

Microcalcification clusters are often an important indicator for the detection of malignancy in mammograms. In many cases, microcalcifications are the only indication of a malignancy. However, the detection of microcalcifications can be a difficult process. They are small and can be embedded in dense tissue. This paper presents a method for automatically detecting microcalcifications. We utilize a high-boost filter to suppress background clutter enabling segmentation even in very dense breast tissue. We then use a threshholding and region growing technique to extract candidate microcalcifications. Likely microcalcifications are then identified by a linear classifier. We apply this method to images selected from the LLNL/UCSF Digital Mammogram Library, and produce a receiver operating characteristic (ROC) curves to detail the trade-off between probability of detection and false alarms. Finally, we exam the ability to properly select a threshold to achieve a desired probability of detection based upon a training set. This is a US government work. There are no restrictions on its use.

Mammographic percent density (PD%) is known to be a strong risk factor for breast cancer. Recent studies also suggest that parenchymal texture features, which are more granular descriptors of the parenchymal pattern, can provide additional information about breast cancer risk. To date, most studies have measured mammographic texture within selected regions of interest (ROIs) in the breast, which cannot adequately capture the complexity of the parenchymal pattern throughout the whole breast. To better characterize patterns of the parenchymal tissue, the authors have developed a fully automated software pipeline based on a novel lattice-based strategy to extract a range of parenchymal texture features from the entire breast region. Digital mammograms from 106 cases with 318 age-matched controls were retrospectively analyzed. The lattice-based approach is based on a regular grid virtually overlaid on each mammographic image. Texture features are computed from the intersection (i.e., lattice) points of the grid lines within the breast, using a local window centered at each lattice point. Using this strategy, a range of statistical (gray-level histogram, co-occurrence, and run-length) and structural (edge-enhancing, local binary pattern, and fractal dimension) features are extracted. To cover the entire breast, the size of the local window for feature extraction is set equal to the lattice grid spacing and optimized experimentally by evaluating different windows sizes. The association between their lattice-based texture features and breast cancer was evaluated using logistic regression with leave-one-out cross validation and further compared to that of breast PD% and commonly used single-ROI texture features extracted from the retroareolar or the central breast region. Classification performance was evaluated using the area under the curve (AUC) of the receiver operating characteristic (ROC). DeLong's test was used to compare the different ROCs in terms of AUC performance. The average univariate performance of the lattice-based features is higher when extracted from smaller than larger window sizes. While not every individual texture feature is superior to breast PD% (AUC: 0.59, STD: 0.03), their combination in multivariate analysis has significantly better performance (AUC: 0.85, STD: 0.02, p < 0.001). The lattice-based texture features also outperform the single-ROI texture features when extracted from the retroareolar or the central breast region (AUC: 0.60-0.74, STD: 0.03). Adding breast PD% does not make a significant performance improvement to the lattice-based texture features or the single-ROI features (p > 0.05). The proposed lattice-based strategy for mammographic texture analysis enables to characterize the parenchymal pattern over the entire breast. As such, these features provide richer information compared to currently used descriptors and may ultimately improve breast cancer risk assessment. Larger studies are warranted to validate these findings and also compare to standard demographic and reproductive risk factors.

Clustered microcalcifications (MCs) in mammograms are an important early sign of breast cancer in women. Their accurate detection is important in computer-aided detection (CADe). In this paper, we integrated the possibilistic fuzzy c-means (PFCM) clustering algorithm and weighted support vector machine (WSVM) for the detection of MC clusters in full-field digital mammograms (FFDM). For each image, suspicious MC regions are extracted with region growing and active contour segmentation. Then geometry and texture features are extracted for each suspicious MC, a mutual information-based supervised criterion is used to select important features, and PFCM is applied to cluster the samples into two clusters. Weights of the samples are calculated based on possibilities and typicality values from the PFCM, and the ground truth labels. A weighted nonlinear SVM is trained. During the test process, when an unknown image is presented, suspicious regions are located with the segmentation step, selected features are extracted, and the suspicious MC regions are classified as containing MC or not by the trained weighted nonlinear SVM. Finally, the MC regions are analyzed with spatial information to locate MC clusters. The proposed method is evaluated using a database of 410 clinical mammograms and compared with a standard unweighted support vector machine (SVM) classifier. The detection performance is evaluated using response receiver operating (ROC) curves and free-response receiver operating characteristic (FROC) curves. The proposed method obtained an area under the ROC curve of 0.8676, while the standard SVM obtained an area of 0.8268 for MC detection. For MC cluster detection, the proposed method obtained a high sensitivity of 92 % with a false-positive rate of 2.3 clusters/image, and it is also better than standard SVM with 4.7 false-positive clusters/image at the same sensitivity.

We propose computer-aided detection (CADe) algorithm for microcalcification (MC) clusters in reconstructed digital breast tomosynthesis (DBT) images. The algorithm consists of prescreening, MC detection, clustering, and false-positive (FP) reduction steps. The DBT images containing the MC-like objects were enhanced by a multiscale Hessian-based three-dimensional (3D) objectness response function and a connected-component segmentation method was applied to extract the cluster seed objects as potential clustering centers of MCs. Secondly, a signal-to-noise ratio (SNR) enhanced image was also generated to detect the individual MC candidates and prescreen the MC-like objects. Each cluster seed candidate was prescreened by counting neighboring individual MC candidates nearby the cluster seed object according to several microcalcification clustering criteria. As a second step, we introduced bounding boxes for the accepted seed candidate, clustered all the overlapping cubes, and examined. After the FP reduction step, the average number of FPs per case was estimated to be 2.47 per DBT volume with a sensitivity of 83.3%.

We are developing computer-aided diagnosis (CAD) schemes for the detection of clustered microcalcifications and masses in digital mammograms. Here, CAD refers to a diagnosis made by a radiologist who uses the computerized analyses of radiographic images as a 'second opinion'. The radiologist would make the final diagnostic decision. The aim of CAD is to improve diagnostic accuracy by reducing the number of missed diagnoses. In this preliminary evaluation, 30 clinical cases from December 1991 having a focal mammographic finding were analyzed.

Previous studies in which the JNDmetrix visual discrimination model (VDM) was applied to predict effects of image display and processing factors on lesion detectability have shown promising results for mammographic images with microcalcification clusters. In those studies, just-noticeable-difference (JND) metrics were computed for signal-present and signal-absent image pairs with the same background. When this method was applied to Gaussian signals in 1/f3 filtered noise, however, it was unable to predict detection thresholds measured in 2AFC trials for different backgrounds. We suggested previously (SPIE 2002) that a statistical model observer using channel responses from single-ended VDM simulations could predict detection performance with different backgrounds. The implementation and evaluation of that VDM-channelized model observer is described in this paper. Model performance was computed for sets of signal and noise images from two observer performance studies involving the detection of simulated or real breast masses. For the first study, the VDM-channelized model observer was able to predict the dependence of detection thresholds on signal size (contrast-detail slope) for 2AFC detection of Gaussian signals on different 1/f3 noise backgrounds. Variations in the detectability of masses in mammograms from the second study correlated well with model performance as a function of display type (LCD vs. CRT) and viewing angle (on-axis vs. 45° off-axis). The performance of the VDM-channelized model observer was superior to results obtained using either the VDM paired discriminability method or a conventional nonprewhitening model observer.