Digital mammography: observer performance study of the effects of pixel size on the characterization of malignant and benign microcalcifications.

RATIONALE AND OBJECTIVES: The authors performed this study to evaluate the effects of pixel size on the characterization of mammographic microcalcifications by radiologists. MATERIALS AND METHODS: Two-view mammograms of 112 microcalcification clusters were digitized with a laser scanner at a pixel size of 35 microm. Images with pixel sizes of 70, 105, and 140 microm were derived from the 35-microm-pixel size images by averaging neighboring pixels. The malignancy or benignity of the microcalcifications had been determined with findings at biopsy or 2-year follow-up. Region-of-interest images containing the microcalcifications were printed with a laser imager. Seven radiologists participated in a receiver operating characteristic (ROC) study to estimate the likelihood of malignancy. The classification accuracy was quantified with the area under the ROC curve (Az). The statistical significance of the differences in the Az values for different pixel sizes was estimated with the Dorfman-Berbaum-Metz method and the Student paired t test. The variance components were analyzed with a bootstrap method. RESULTS: The higher-resolution images did not result in better classification; the average Az with a pixel size of 35 microm was lower than that with pixel sizes of 70 and 105 microm. The differences in Az between different pixel sizes did not achieve statistical significance. CONCLUSION: Pixel sizes in the range studied do not have a strong effect on radiologists' accuracy in the characterization of microcalcifications. The low specificity of the image features of microcalcifications and the large interobserver and intraobserver variabilities may have prevented small advantages in image resolution from being observed.

Follow-up of benign results of stereotactic core breast biopsy.

RATIONALE AND OBJECTIVES: The purpose of this study was to obtain long-term follow-up data on women with benign histologic results of a breast stereotactic core needle biopsy (CNB). MATERIALS AND METHODS: Mammography charts of 300 consecutive women who underwent prone stereotactic CNB with digital radiography were reviewed. Women with frankly malignant or suspicious histologic findings (51 patients) or a technically unsuccessful stereotactic CNB (one patient) were excluded. The remaining 248 benign core biopsies in 229 women were included in the study. RESULTS: Follow-up mammograms were obtained for 152 lesions with benign histologic results following stereotactic CNB. The mean length of follow-up after stereotactic CNB was 34.6 months. Cancer was diagnosed in six women who underwent surgical biopsies 1/2 to 30 months after benign stereotactic CNB. An initial chart review demonstrated that no follow-up data were available for 64 lesions, and information was missing for an additional seven. CONCLUSION: SCNB remains a sampling procedure that can result in false-negative histologic results. Intrinsic procedural issues were identified that could minimize the potential for missing a malignancy. Goals for patient compliance with follow-up recommendations fell short of expectations.

Improvement of radiologists' characterization of mammographic masses by using computer-aided diagnosis: an ROC study.

PURPOSE: To evaluate the effects of computer-aided diagnosis (CAD) on radiologists' classification of malignant and benign masses seen on mammograms. MATERIALS AND METHODS: The authors previously developed an automated computer program for estimation of the relative malignancy rating of masses. In the present study, the authors conducted observer performance experiments with receiver operating characteristic (ROC) methodology to evaluate the effects of computer estimates on radiologists' confidence ratings. Six radiologists assessed biopsy-proved masses with and without CAD. Two experiments, one with a single view and the other with two views, were conducted. The classification accuracy was quantified by using the area under the ROC curve, Az. RESULTS: For the reading of 238 images, the Az value for the computer classifier was 0.92. The radiologists' Az values ranged from 0.79 to 0.92 without CAD and improved to 0.87-0.96 with CAD. For the reading of a subset of 76 paired views, the radiologists' Az values ranged from 0.88 to 0.95 without CAD and improved to 0.93-0.97 with CAD. Improvements in the reading of the two sets of images were statistically significant (P = .022 and .007, respectively). An improved positive predictive value as a function of the false-negative fraction was predicted from the improved ROC curves. CONCLUSION: CAD may be useful for assisting radiologists in classification of masses and thereby potentially help reduce unnecessary biopsies.

Computerized analysis of mammographic microcalcifications in morphological and texture feature spaces.

We are developing computerized feature extraction and classification methods to analyze malignant and benign microcalcifications on digitized mammograms. Morphological features that described the size, contrast, and shape of microcalcifications and their variations within a cluster were designed to characterize microcalcifications segmented from the mammographic background. Texture features were derived from the spatial gray-level dependence (SGLD) matrices constructed at multiple distances and directions from tissue regions containing microcalcifications. A genetic algorithm (GA) based feature selection technique was used to select the best feature subset from the multi-dimensional feature spaces. The GA-based method was compared to the commonly used feature selection method based on the stepwise linear discriminant analysis (LDA) procedure. Linear discriminant classifiers using the selected features as input predictor variables were formulated for the classification task. The discriminant scores output from the classifiers were analyzed by receiver operating characteristic (ROC) methodology and the classification accuracy was quantified by the area, Az, under the ROC curve. We analyzed a data set of 145 mammographic microcalcification clusters in this study. It was found that the feature subsets selected by the GA-based method are comparable to or slightly better than those selected by the stepwise LDA method. The texture features (Az = 0.84) were more effective than morphological features (Az = 0.79) in distinguishing malignant and benign microcalcifications. The highest classification accuracy (Az = 0.89) was obtained in the combined texture and morphological feature space. The improvement was statistically significant in comparison to classification in either the morphological (p = 0.002) or the texture (p = 0.04) feature space alone. The classifier using the best feature subset from the combined feature space and an appropriate decision threshold could correctly identify 35% of the benign clusters without missing a malignant cluster. When the average discriminant score from all views of the same cluster was used for classification, the Az value increased to 0.93 and the classifier could identify 50% of the benign clusters at 100% sensitivity for malignancy. Alternatively, if the minimum discriminant score from all views of the same cluster was used, the Az value would be 0.90 and a specificity of 32% would be obtained at 100% sensitivity. The results of this study indicate the potential of using combined morphological and texture features for computer-aided classification of microcalcifications.

Mammographic appearance of the breasts during pregnancy and lactation: false assumptions.

RATIONALE AND OBJECTIVES: It is believed that pregnant and lactating women have dense breasts, thereby limiting the usefulness of mammography. To our knowledge, no reports have been published on this topic for nearly 4 decades. The purpose of our study was to determine whether this assumption is accurate given current state-of-the-art mammography. MATERIALS AND METHODS: Mammograms of 18 women (six pregnant, seven lactating, and five who recently discontinued lactation) were examined and compared with their baseline (before pregnancy or lactation) mammograms and with mammograms of 18 age-matched control subjects. Studies were scored for breast density according to the Breast Imaging Reporting and Data System and evaluated for change in density and size. RESULTS: Three of the six pregnant women had unchanged breast density compared with baseline studies and had scattered fibroglandular or heterogeneously dense tissue. Of the three without baseline studies, one had extremely dense, one had heterogeneously dense, and one had scattered fibroglandular tissue. All seven lactating women had either heterogeneously dense or extremely dense tissue. The breast tissue in four was unchanged in density and increased in two; no baseline study was available for the remaining patient. Seven studies in five women who had discontinued lactation 1 week to 5 months prior to mammography showed no change in density compared with baseline. CONCLUSION: Pregnant and lactating women do not always have dense breasts, and mammography can be performed without substantial concern for the limitations of breast density. Mammography can be as useful in these women as it is in other women with breast signs and symptoms.

False-positive reduction technique for detection of masses on digital mammograms: global and local multiresolution texture analysis.

We investigated the application of multiresolution global and local texture features to reduce false-positive detection in a computerized mass detection program. One hundred and sixty-eight digitized mammograms were randomly and equally divided into training and test groups. From these mammograms, two datasets were formed. The first dataset (manual) contained four regions of interest (ROIs) selected manually from each of the mammograms. One of the four ROIs contained a biopsy-proven mass and the other three contained normal parenchyma, including dense, mixed dense/fatty, and fatty tissues. The second dataset (hybrid) contained the manually extracted mass ROIs, along with normal tissue ROIs extracted by an automated Density-Weighted Contrast Enhancement (DWCE) algorithm as false-positive detections. A wavelet transform was used to decompose an ROI into several scales. Global texture features were derived from the low-pass coefficients in the wavelet transformed images. Local texture features were calculated from the suspicious object and the peripheral subregions. Linear discriminant models using effective features selected from the global, local, or combined feature spaces were established to maximize the separation between masses and normal tissue. Receiver Operating Characteristic (ROC) analysis was conducted to evaluate the classifier performance. The classification accuracy using global features were comparable to that using local features. With both global and local features, the average area, Az, under the test ROC curve, reached 0.92 for the manual dataset and 0.96 for the hybrid dataset, demonstrating statistically significant improvement over those obtained with global or local features alone. The results indicated the effectiveness of the combined global and local features in the classification of masses and normal tissue for false-positive reduction.

Computerized classification of malignant and benign microcalcifications on mammograms: texture analysis using an artificial neural network.

We investigated the feasibility of using texture features extracted from mammograms to predict whether the presence of microcalcifications is associated with malignant or benign pathology. Eighty-six mammograms from 54 cases (26 benign and 28 malignant) were used as case samples. All lesions had been recommended for surgical biopsy by specialists in breast imaging. A region of interest (ROI) containing the microcalcifications was first corrected for the low-frequency background density variation. Spatial grey level dependence (SGLD) matrices at ten different pixel distances in both the axial and diagonal directions were constructed from the background-corrected ROI. Thirteen texture measures were extracted from each SGLD matrix. Using a stepwise feature selection technique, which maximized the separation of the two class distributions, subsets of texture features were selected from the multi-dimensional feature space. A backpropagation artificial neural network (ANN) classifier was trained and tested with a leave-one-case-out method to recognize the malignant or benign microcalcification clusters. The performance of the ANN was analysed with receiver operating characteristic (ROC) methodology. It was found that a subset of six texture features provided the highest classification accuracy among the feature sets studied. The ANN classifier achieved an area under the ROC curve of 0.88. By setting an appropriate decision threshold, 11 of the 28 benign cases were correctly identified (39% specificity) without missing any malignant cases (100% sensitivity) for patients who had undergone biopsy. This preliminary result indicates that computerized texture analysis can extract mammographic information that is not apparent by visual inspection. The computer-extracted texture information may be used to assist in mammographic interpretation, with the potential to reduce biopsies of benign cases and improve the positive predictive value of mammography.

The 3D and 2D color flow display of breast masses.

A prospective study was performed in 24 women with breast masses on mammography going on to surgical biopsy. 2D and 3D power mode and frequency shift color flow Doppler scanning and display were compared. Vessels were displayed as rotatable color volumes in 3D, superimposed on gray-scale slices. The latter were stepped sequentially through the imaged volume. Radiologists rated the masses in each display (3D, 2D and videotapes) on a scale of 1 to 5 (5 = most suspicious) for each of six conventional gray-scale and six new vascular criteria. Thirteen masses proved to be benign and 11 were malignant. 3D provided a stronger subjective appreciation of vascular morphology and allowed somewhat better ultrasound discrimination of malignant masses than did the 2D images or videotapes (specificities of 85%, 79% and 71%, respectively, at a sensitivity of 90%). Only in 3D did the vascularity measures display a trend towards significance in this small study.

Image feature selection by a genetic algorithm: application to classification of mass and normal breast tissue.

We investigated a new approach to feature selection, and demonstrated its application in the task of differentiating regions of interest (ROIs) on mammograms as either mass or normal tissue. The classifier included a genetic algorithm (GA) for image feature selection, and a linear discriminant classifier or a backpropagation neural network (BPN) for formulation of the classifier outputs. The GA-based feature selection was guided by higher probabilities of survival for fitter combinations of features, where the fitness measure was the area Az under the receiver operating characteristic (ROC) curve. We studied the effect of different GA parameters on classification accuracy, and compared the results to those obtained with stepwise feature selection. The data set used in this study consisted of 168 ROIs containing biopsy-proven masses and 504 ROIs containing normal tissue. From each ROI, a total of 587 features were extracted, of which 572 were texture features and 15 were morphological features. The GA was trained and tested with several different partitionings of the ROIs into training and testing sets. With the best combination of the GA parameters, the average test Az value using a linear discriminant classifier reached 0.90, as compared to 0.89 for stepwise feature selection. Test Az values with a BPN classifier and a more limited feature pool were 0.90 with GA-based feature selection, and 0.89 for stepwise feature selection. The use of a GA in tailoring classifiers with specific design characteristics was also discussed. This study indicates that a GA can provide versatility in the design of linear or nonlinear classifiers without a trade-off in the effectiveness of the selected features.

Automated detection of breast masses on mammograms using adaptive contrast enhancement and texture classification.

This paper presents segmentation and classification results of an automated algorithm for the detection of breast masses on digitized mammograms. Potential mass regions were first identified using density-weighted contrast enhancement (DWCE) segmentation applied to single-view mammograms. Once the potential mass regions had been identified, multiresolution texture features extracted from wavelet coefficients were calculated, and linear discriminant analysis (LDA) was used to classify the regions as breast masses or normal tissue. In this article the overall detection results for two independent sets of 84 mammograms used alternately for training and test were evaluated by free-response receiver operating characteristics (FROC) analysis. The test results indicate that this new algorithm produced approximately 4.4 false positive per image at a true positive detection rate of 90% and 2.3 false positives per image at a true positive rate of 80%.

Locally advanced breast carcinoma: accuracy of mammography versus clinical examination in the prediction of residual disease after chemotherapy.

PURPOSE: To determine the mammographic features of locally advanced breast carcinoma treated with neoadjuvant chemotherapy and to evaluate the accuracy of mammography in the prediction of residual carcinoma. MATERIALS AND METHODS: Of 90 women treated with hormonally synchronized cytotoxic therapy before mastectomy or lumpectomy for advanced breast carcinoma, 56 were selected because they had undergone mammography before and after neo-adjuvant therapy. Mammographic and clinical opinion on the presence of residual disease was compared with histologic results. RESULTS: Fifty-four (96%) of 56 women had a complete (n = 34 [61%]) or partial (n = 20 [36%]) clinical response. Thirteen (23%) of 56 women had no residual tumor. Sensitivity of mammography in the prediction of residual carcinoma was greater than that of clinical examination (79% vs 49%), but specificity was lower (77% vs 92%). In 24 women with inflammatory carcinoma, sensitivity of mammography was 78% while that of clinical examination was 39%; specificity was equal (83%). CONCLUSION: Mammography was more sensitive than clinical examination in the prediction of residual carcinoma; it was not accurate enough to obviate surgical biopsy.

Classification of mass and normal breast tissue: a convolution neural network classifier with spatial domain and texture images.

The authors investigated the classification of regions of interest (ROI's) on mammograms as either mass or normal tissue using a convolution neural network (CNN). A CNN is a backpropagation neural network with two-dimensional (2-D) weight kernels that operate on images. A generalized, fast and stable implementation of the CNN was developed. The input images to the CNN were obtained from the ROI's using two techniques. The first technique employed averaging and subsampling. The second technique employed texture feature extraction methods applied to small subregions inside the ROI. Features computed over different subregions were arranged as texture images, which were subsequently used as CNN inputs. The effects of CNN architecture and texture feature parameters on classification accuracy were studied. Receiver operating characteristic (ROC) methodology was used to evaluate the classification accuracy. A data set consisting of 168 ROIs containing biopsy-proven masses and 504 ROI's containing normal breast tissue was extracted from 168 mammograms by radiologists experienced in mammography. This data set was used for training and testing the CNN. With the best combination of CNN architecture and texture feature parameters, the area under the test ROC curve reached 0.87, which corresponded to a true-positive fraction of 90% at a false positive fraction of 31%. The authors' results demonstrate the feasibility of using a CNN for classification of masses and normal tissue on mammograms.

Classification of mass and normal breast tissue on digital mammograms: multiresolution texture analysis.

We investigated the feasibility of using multiresolution texture analysis for differentiation of masses from normal breast tissue on mammograms. The wavelet transform was used to decompose regions of interest (ROIs) on digitized mammograms into several scales. Multiresolution texture features were calculated from the spatial gray level dependence matrices of (1) the original images at variable distances between the pixel pairs, (2) the wavelet coefficients at different scales, and (3) the wavelet coefficients up to certain scale and then at variable distances between the pixel pairs. In this study, 168 ROIs containing biopsy-proven masses and 504 ROIs containing normal parenchyma were used as the data set. The mass ROIs were randomly and equally divided into training and test groups along with corresponding normal ROIs from the same film. Stepwise linear discriminant analysis was used to select optimal features from the multiresolution texture feature space to maximize the separation of mass and normal tissue for all ROIs. We found that texture features at large pixel distances are important for the classification task. The wavelet transform can effectively condense the image information into its coefficients. With texture features based on the wavelet coefficients and variable distances, the area Az under the receiver operating characteristic curve reached 0.89 and 0.86 for the training and test groups, respectively. The results demonstrate that a linear discriminant classifier using the multiresolution texture features can effectively classify masses from normal tissue on mammograms.

Dynamic three-dimensional imaging with partial k-space sampling: initial application for gadolinium-enhanced rate characterization of breast lesions.

PURPOSE: To evaluate a method to monitor gadolinium enhancement patterns at magnetic resonance (MR) imaging with high temporal resolution and full coverage through both breasts. MATERIALS AND METHODS: In 12 patients with 13 masses, including nine carcinoma, nonenhanced three-dimensional MR imaging was performed with full-matrix resolution. At dynamic imaging, 32 serial passes were made during bolus administration of contrast material, and temporal resolution was reduced to 12 seconds by collecting the central (low spatial frequency) 32 x 16 or 16 x 16 phase-encode views. Full-matrix dynamic images were reconstructed by complementing central phase-encode data with precontrast data from peripheral high-spatial-frequency views. RESULTS: Results at time-course analysis with a mono-exponential saturation model indicated malignant lesions tend to show rapid (< 60 seconds) contrast change relative to benign masses and normal tissues. One cancer displayed an exceptionally slow contrast change (260 seconds). CONCLUSION: The technical objectives of full tissue coverage, rapid temporal sampling, and quantification of enhancement curves are met with this method for certain lesions (> 5 mm in largest diameter).

Computer-aided classification of mammographic masses and normal tissue: linear discriminant analysis in texture feature space.

We studied the effectiveness of using texture features derived from spatial grey level dependence (SGLD) matrices for classification of masses and normal breast tissue on mammograms. One hundred and sixty-eight regions of interest (ROIS) containing biopsy-proven masses and 504 ROIS containing normal breast tissue were extracted from digitized mammograms for this study. Eight features were calculated for each ROI. The importance of each feature in distinguishing masses from normal tissue was determined by stepwise linear discriminant analysis. Receiver operating characteristic (ROC) methodology was used to evaluate the classification accuracy. We investigated the dependence of classification accuracy on the input features, and on the pixel distance and bit depth in the construction of the SGLD matrices. It was found that five of the texture features were important for the classification. The dependence of classification accuracy on distance and bit depth was weak for distances greater than 12 pixels and bit depths greater than seven bits. By randomly and equally dividing the data set into two groups, the classifier was trained and tested on independent data sets. The classifier achieved an average area under the ROC curve, Az, of 0.84 during training and 0.82 during testing. The results demonstrate the feasibility of using linear discriminant analysis in the texture feature space for classification of true and false detections of masses on mammograms in a computer-aided diagnosis scheme.

Invasive lobular carcinoma: sonographic appearance and role of sonography in improving diagnostic sensitivity.

PURPOSE: To characterize the ultrasonographic (US) appearance of invasive lobular carcinoma (ILC) and to assess the potential role of US in the earlier detection of ILC. MATERIALS AND METHODS: US scans in 19 patients with ILC were retrospectively studied for the presence of a mass, characteristics of the margins, internal echogenicity, and attenuation effects. RESULTS: US showed masses in 13 of the 19 patients (68% sensitivity). Irregularly marginated masses with heterogeneous internal echoes and acoustic attenuation were present in seven patients. A variety of US findings, mimicking a benign lesion, were noted in the other six patients. US sensitivity in the detection of small cancers (< 1 cm) was 25% (one of four patients). Mammographic sensitivity in the detection of ILC in this series was 89% (17 of 19 patients). CONCLUSION: ILC has a variety of US appearances. US was insensitive and nonspecific in the diagnosis of ILC, especially for small cancers. A negative US result should not deter surgical biopsy if indicated by mammographic findings or clinical findings.

New methods for imaging the breast: techniques, findings, and potential.

Mammography is currently the best imaging technique for the early detection and diagnosis of breast cancer. Although numerous advances and improvements in mammography in the past decades have greatly improved image quality, the technique is not without shortcomings that limit its sensitivity and specificity. Multiple areas of research have therefore been sought not only to improve film/screen mammography, but also to consider entirely new techniques in the study of breast cancer. Although this review is not intended to include all methods currently under investigation, those chosen for discussion represent areas where major efforts have provided data that suggest exciting future applications. These include MR imaging, digital mammography, computer-aided diagnosis (CAD), positron-emission tomography (PET), and single-photon emission planar CT imaging (SPECT).

Breast thickness in routine mammograms: effect on image quality and radiation dose.

OBJECTIVE: The purpose of this study was to compare the thickness of the compressed breast between mediolateral oblique and craniocaudal mammograms and to relate these differences in thickness to image quality and radiation dose. These differences may partially explain why some subtle tumors are better visualized on the craniocaudal view. SUBJECTS AND METHODS: The study population consisted of 250 paired mediolateral oblique and craniocaudal mammograms obtained on one mammographic unit by seven certified mammography technologists during a 2-month period. Only women with breast implants, prior lumpectomy and radiotherapy, or chest wall deformity were excluded. The digital readout of compressed breast thickness and applied compression force was recorded. Mammographic positioning was assessed using standard criteria. Absorbed radiation dose at different thicknesses was measured with a BR-12 breast phantom. Image quality differences for geometric unsharpness and contrast were calculated for the observed breast thickness differences between mediolateral oblique and craniocaudal mammograms. RESULTS: The mean thickness of the compressed breast on the craniocaudal view was less than the mean thickness on the mediolateral oblique view (4.4 versus 4.8 cm, p < .0001) despite the greater force used to compress the breast for mediolateral oblique than for craniocaudal views (93 versus 86 newtons, p < .0001). The breast thickness on the mediolateral oblique view exceeded that on the craniocaudal view in 98 (84%) of 117 pairs that differed in thickness by 5 mm or more and 46 (94%) of 49 pairs that differed by 10 mm or more (p < .0001). Geometric unsharpness increased by 8% and 19% when a 4.4-cm-thick breast was compared to a 4.8- and 5.4-cm-thick breast, respectively. A 5% and 12% loss of contrast was noted when a 4.4-cm-thick breast was compared to a 4.8- and 5.4-cm-thick breast. Mean glandular radiation dose at 4.4, 4.8, and 5.4 cm was 1.40, 1.70, and 2.33 mGy, respectively. CONCLUSION: The compressed breast is 8% thicker on mediolateral oblique than on craniocaudal mammograms, a small but statistically significant difference. This difference results in a small loss of spatial and contrast resolution on the mediolateral oblique views and an increase in radiation dose. These image quality differences may partially explain why some subtle carcinomas are better visualized on the craniocaudal view.

Computer-aided diagnosis in mammography: classification of mass and normal tissue by texture analysis.

Computer-aided diagnosis schemes are being developed to assist radiologists in mammographic interpretation. In this study, we investigated whether texture features could be used to distinguish between mass and non-mass regions in clinical mammograms. Forty-five regions of interest (ROIs) containing true masses with various degrees of visibility and 135 ROIs containing normal breast parenchyma were extracted manually from digitized mammograms as case samples. Spatial-grey-level-dependence (SGLD) matrices of each ROI were calculated and eight texture features were calculated from the SGLD matrices. The correlation and class-distance properties of extracted texture features were analysed. Selected texture features were input into a modified decision-tree classification scheme. The performance of the classifier was evaluated for different feature combinations and orders of features on the tree. A classification accuracy of about 89% sensitivity and 76% specificity was obtained for ordered features, sum average, correlation, and energy, during the training procedure. With a leave-one-out method, the test result was about 76% sensitivity and 64% specificity. The results of this preliminary study demonstrate the feasibility of using texture information for classification of mass and normal breast tissue, which will be likely to be useful for classifying true and false detections in computer-aided diagnosis programmes.