Search

We compared the performance of different Deep learning-convolutional neural network (DL-CNN) models for bladder cancer treatment response assessment based on transfer learning by freezing different DL-CNN layers and varying the DL-CNN structure. Pre- and posttreatment computed tomography scans of 123 patients (cancers, 129; pre- and posttreatment cancer pairs, 158) undergoing chemotherapy were collected. After chemotherapy 33% of patients had T0 stage cancer (complete response). Regions of interest in pre- and posttreatment scans were extracted from the segmented lesions and combined into hybrid pre -post image pairs (h-ROIs). Training (pairs, 94; h-ROIs, 6209), validation (10 pairs) and test sets (54 pairs) were obtained. The DL-CNN consisted of 2 convolution (C1-C2), 2 locally connected (L3-L4), and 1 fully connected layers. The DL-CNN was trained with h-ROIs to classify cancers as fully responding (stage T0) or not fully responding to chemotherapy. Two radiologists provided lesion likelihood of being stage T0 posttreatment. The test area under the ROC curve (AUC) was 0.73 for T0 prediction by the base DL-CNN structure with randomly initialized weights. The base DL-CNN structure with pretrained weights and transfer learning (no frozen layers) achieved test AUC of 0.79. The test AUCs for 3 modified DL-CNN structures (different C1-C2 max pooling filter sizes, strides, and padding, with transfer learning) were 0.72, 0.86, and 0.69. For the base DL-CNN with (C1) frozen, (C1-C2) frozen, and (C1-C2-L3) frozen, the test AUCs were 0.81, 0.78, and 0.71, respectively. The radiologists' AUCs were 0.76 and 0.77. DL-CNN performed better with pretrained than randomly initialized weights.

Purpose: To evaluate the feasibility of using an objective computer-aided system to assess bladder cancer stage in CT Urography (CTU)., Materials and Methods: A dataset consisting of 84 bladder cancer lesions from 76 CTU cases was used to develop the computerized system for bladder cancer staging based on machine learning approaches. The cases were grouped into two classes based on pathological stage ≥ T2 or below T2, which is the decision threshold for neoadjuvant chemotherapy treatment clinically. There were 43 cancers below stage T2 and 41 cancers at stage T2 or above. All 84 lesions were automatically segmented using our previously developed auto-initialized cascaded level sets (AI-CALS) method. Morphological and texture features were extracted. The features were divided into subspaces of morphological features only, texture features only, and a combined set of both morphological and texture features. The dataset was split into Set 1 and Set 2 for two-fold cross-validation. Stepwise feature selection was used to select the most effective features. A linear discriminant analysis (LDA), a neural network (NN), a support vector machine (SVM), and a random forest (RAF) classifier were used to combine the features into a single score. The classification accuracy of the four classifiers was compared using the area under the receiver operating characteristic (ROC) curve (A z )., Results: Based on the texture features only, the LDA classifier achieved a test A z of 0.91 on Set 1 and a test A z of 0.88 on Set 2. The test A z of the NN classifier for Set 1 and Set 2 were 0.89 and 0.92, respectively. The SVM classifier achieved test A z of 0.91 on Set 1 and test A z of 0.89 on Set 2. The test A z of the RAF classifier for Set 1 and Set 2 was 0.89 and 0.97, respectively. The morphological features alone, the texture features alone, and the combined feature set achieved comparable classification performance., Conclusion: The predictive model developed in this study shows promise as a classification tool for stratifying bladder cancer into two staging categories: greater than or equal to stage T2 and below stage T2., (© 2017 American Association of Physicists in Medicine.)

Rationale and Objectives: This study aimed to compare Breast Imaging Reporting and Data System (BI-RADS) assessment of lesions in two-view digital mammogram (DM) to two-view wide-angle digital breast tomosynthesis (DBT) without DM., Materials and Methods: With Institutional Review Board approval and written informed consent, two-view DBTs were acquired from 134 subjects and the corresponding DMs were collected retrospectively. The study included 125 subjects with 61 malignant (size: 3.9-36.9 mm, median: 13.4 mm) and 81 benign lesions (size: 4.8-43.8 mm, median: 12.0 mm), and 9 normal subjects. The cases in the two modalities were read independently by six experienced Mammography Quality Standards Act radiologists in a fully crossed counterbalanced manner. The readers were blinded to the prevalence of malignant, benign, or normal cases and were asked to assess the lesions based on the BI-RADS lexicon. The ratings were analyzed by the receiver operating characteristic methodology., Results: Lesion conspicuity was significantly higher (P << .0001) and fewer lesion margins were considered obscured in DBT. The mean area under the receiver operating characteristic curve for the six readers increased significantly (P = .0001) from 0.783 (range: 0.723-0.886) for DM to 0.911 (range: 0.884-0.936) for DBT. Of the 366 ratings for malignant lesions, 343 on DBT and 278 on DM were rated as BI-RADS 4a and above. Of the 486 ratings for benign lesions, 220 on DBT and 206 on DM were rated as BI-RADS 4a and above. On average, 17.8% (65 of 366) more malignant lesions and 2.9% (14 of 486) more benign lesions would be recommended for biopsy using DBT. The inter-radiologist variability was reduced significantly., Conclusion: With DBT alone, the BI-RADS assessment of breast lesions and inter-radiologist reliability were significantly improved compared to DM., (Copyright © 2017 The Association of University Radiologists. Published by Elsevier Inc. All rights reserved.)

Cross-sectional X-ray imaging has become the standard for staging most solid organ malignancies. However, for some malignancies such as urinary bladder cancer, the ability to accurately assess local extent of the disease and understand response to systemic chemotherapy is limited with current imaging approaches. In this study, we explored the feasibility that radiomics-based predictive models using pre- and post-treatment computed tomography (CT) images might be able to distinguish between bladder cancers with and without complete chemotherapy responses. We assessed three unique radiomics-based predictive models, each of which employed different fundamental design principles ranging from a pattern recognition method via deep-learning convolution neural network (DL-CNN), to a more deterministic radiomics feature-based approach and then a bridging method between the two, utilizing a system which extracts radiomics features from the image patterns. Our study indicates that the computerized assessment using radiomics information from the pre- and post-treatment CT of bladder cancer patients has the potential to assist in assessment of treatment response.

Assessing the response of bladder cancer to neoadjuvant chemotherapy is crucial for reducing morbidity and increasing quality of life of patients. Changes in tumor volume during treatment is generally used to predict treatment outcome. We are developing a method for bladder cancer segmentation in CT using a pilot data set of 62 cases. 65 000 regions of interests were extracted from pre-treatment CT images to train a deep-learning convolution neural network (DL-CNN) for tumor boundary detection using leave-one-case-out cross-validation. The results were compared to our previous AI-CALS method. For all lesions in the data set, the longest diameter and its perpendicular were measured by two radiologists, and 3D manual segmentation was obtained from one radiologist. The World Health Organization (WHO) criteria and the Response Evaluation Criteria In Solid Tumors (RECIST) were calculated, and the prediction accuracy of complete response to chemotherapy was estimated by the area under the receiver operating characteristic curve (AUC). The AUCs were 0.73 ± 0.06, 0.70 ± 0.07, and 0.70 ± 0.06, respectively, for the volume change calculated using DL-CNN segmentation, the AI-CALS and the manual contours. The differences did not achieve statistical significance. The AUCs using the WHO criteria were 0.63 ± 0.07 and 0.61 ± 0.06, while the AUCs using RECIST were 0.65 ± 007 and 0.63 ± 0.06 for the two radiologists, respectively. Our results indicate that DL-CNN can produce accurate bladder cancer segmentation for calculation of tumor size change in response to treatment. The volume change performed better than the estimations from the WHO criteria and RECIST for the prediction of complete response.

Purpose: To investigate the dependence of microcalcification cluster detectability on tomographic scan angle, angular increment, and number of projection views acquired at digital breast tomosynthesis ( DBT digital breast tomosynthesis )., Materials and Methods: A prototype DBT digital breast tomosynthesis system operated in step-and-shoot mode was used to image breast phantoms. Four 5-cm-thick phantoms embedded with 81 simulated microcalcification clusters of three speck sizes (subtle, medium, and obvious) were imaged by using a rhodium target and rhodium filter with 29 kV, 50 mAs, and seven acquisition protocols. Fixed angular increments were used in four protocols (denoted as scan angle, angular increment, and number of projection views, respectively: 16°, 1°, and 17; 24°, 3°, and nine; 30°, 3°, and 11; and 60°, 3°, and 21), and variable increments were used in three (40°, variable, and 13; 40°, variable, and 15; and 60°, variable, and 21). The reconstructed DBT digital breast tomosynthesis images were interpreted by six radiologists who located the microcalcification clusters and rated their conspicuity., Results: The mean sensitivity for detection of subtle clusters ranged from 80% (22.5 of 28) to 96% (26.8 of 28) for the seven DBT digital breast tomosynthesis protocols; the highest sensitivity was achieved with the 16°, 1°, and 17 protocol (96%), but the difference was significant only for the 60°, 3°, and 21 protocol (80%, P < .002) and did not reach significance for the other five protocols (P = .01-.15). The mean sensitivity for detection of medium and obvious clusters ranged from 97% (28.2 of 29) to 100% (24 of 24), but the differences fell short of significance (P = .08 to >.99). The conspicuity of subtle and medium clusters with the 16°, 1°, and 17 protocol was rated higher than those with other protocols; the differences were significant for subtle clusters with the 24°, 3°, and nine protocol and for medium clusters with 24°, 3°, and nine; 30°, 3°, and 11; 60°, 3° and 21; and 60°, variable, and 21 protocols (P < .002)., Conclusion: With imaging that did not include x-ray source motion or patient motion during acquisition of the projection views, narrow-angle DBT digital breast tomosynthesis provided higher sensitivity and conspicuity than wide-angle DBT digital breast tomosynthesis for subtle microcalcification clusters., (© RSNA, 2014.)

The effect of acquisition geometry in digital breast tomosynthesis was evaluated with studies of contrast-to-noise ratios (CNRs) and observer preference. Contrast-detail (CD) test objects in 5 cm thick phantoms with breast-like backgrounds were imaged. Twelve different angular acquisitions (average glandular dose for each ~1.1 mGy) were performed ranging from narrow angle 16° with 17 projection views (16d17p) to wide angle 64d17p. Focal slices of SART-reconstructed images of the CD arrays were selected for CNR computations and the reader preference study. For the latter, pairs of images obtained with different acquisition geometries were randomized and scored by 7 trained readers. The total scores for all images and readings for each acquisition geometry were compared as were the CNRs. In general, readers preferred images acquired with wide angle as opposed to narrow angle geometries. The mean percent preferred was highly correlated with tomosynthesis angle (R = 0.91). The highest scoring geometries were 60d21p (95%), 64d17p (80%), and 48d17p (72%); the lowest scoring were 16d17p (4%), 24d9p (17%) and 24d13p (33%). The measured CNRs for the various acquisitions showed much overlap but were overall highest for wide-angle acquisitions. Finally, the mean reader scores were well correlated with the mean CNRs (R = 0.83).

The purpose of this study was to evaluate the outcomes and cancer rate in solid palpable masses with benign features assessed as BI-RADS 3 or 4A. This study was Institutional Review Board approved. Mammography and breast ultrasound reports in our Radiology Information System were searched for solid, palpable masses with benign features described from 1/1/2000 to 12/31/2009, and retrospectively reviewed. Those masses prospectively assessed as BI-RADS 3 or 4A, or suggestive of a fibroadenoma or other benign pathology were retrieved. Chart review was used to assess outcomes and cancer rate. Basic summary measures were summarized and compared between BI-RADS 3 and 4A groups using Wilcoxon Rank Sum test for continuous data or Fisher's exact test for categorical data. The cancer rate across age quartiles was assessed using Cochran-Armitage trend test. 573 solid palpable masses with benign features in 487 women were identified. There were 197 BI-RADS 3 and 376 BI-RADS 4A masses. The overall cancer rate was 1.6 % (9/573). All cancers were BI-RADS 4A (cancer rate 2.4 %-9/376). Smaller mean size and younger age at presentation in BI-RADS 3 women was found compared to BI-RADS 4A (P < 0.0001). There was a significant increase in cancer rate across age quartiles (P = 0.03124). The cancer rate is very low in solid palpable masses with benign features. In particular, BI-RADS 3 palpable masses in young women may undergo close surveillance without immediate biopsy, confirming what other investigators have found. All cancers were in the BI-RADS 4A group with increasing incidence with age, with over half occurring in women over 40 years old. Palpable masses in women 40 and older with benign features should be considered for immediate biopsy.

Purpose: We are developing a decision tree content-based image retrieval (DTCBIR) CADx system to assist radiologists in characterization of breast masses on ultrasound images., Methods: Three DTCBIR configurations, including decision tree with boosting (DTb), decision tree with full leaf features (DTL), and decision tree with selected leaf features (DTLs) were compared. For DTb, features of a query mass were combined first into a merged feature score and then masses with similar scores were retrieved. For DTL and DTLs, similar masses were retrieved based on the Euclidean distance between feature vectors of the query and those of selected references. For each DTCBIR configuration, we investigated the use of full feature set and subset of features selected by the stepwise linear discriminant analysis (LDA) and simplex optimization method, resulting in six retrieval methods and selected five, DTb-lda, DTL-lda, DTb-full, DTL-full, and DTLs-full, for the observer study. Three MQSA radiologists rated similarities between the query mass and computer-retrieved three most similar masses using nine-point similarity scale (9 = very similar)., Results: For DTb-lda, DTL-lda, DTb-full, DTL-full, and DTLs-full, average A(z) values were 0.90 ± 0.03, 0.85 ± 0.04, 0.87 ± 0.04, 0.79 ± 0.05, and 0.71 ± 0.06, respectively, and average similarity ratings were 5.00, 5.41, 4.96, 5.33, and 5.13, respectively., Conclusions: The DTL-lda is a promising DTCBIR CADx configuration which had simple tree structure, good classification performance, and highest similarity rating.

Objectives: The purpose of this study was to retrospectively evaluate the effect of 3-dimensional automated ultrasound (3D-AUS) as an adjunct to digital breast tomosynthesis (DBT) on radiologists' performance and confidence in discriminating malignant and benign breast masses., Methods: Two-view DBT (craniocaudal and mediolateral oblique or lateral) and single-view 3D-AUS images were acquired from 51 patients with subsequently biopsy-proven masses (13 malignant and 38 benign). Six experienced radiologists rated, on a 13-point scale, the likelihood of malignancy of an identified mass, first by reading the DBT images alone, followed immediately by reading the DBT images with automatically coregistered 3D-AUS images. The diagnostic performance of each method was measured using receiver operating characteristic (ROC) curve analysis and changes in sensitivity and specificity with the McNemar test. After each reading, radiologists took a survey to rate their confidence level in using DBT alone versus combined DBT/3D-AUS as potential screening modalities., Results: The 6 radiologists had an average area under the ROC curve of 0.92 for both modalities (range, 0.89-0.97 for DBT and 0.90-0.94 for DBT/3D-AUS). With a Breast Imaging Reporting and Data System rating of 4 as the threshold for biopsy recommendation, the average sensitivity of the radiologists increased from 96% to 100% (P > .08) with 3D-AUS, whereas the specificity decreased from 33% to 25% (P > .28). Survey responses indicated increased confidence in potentially using DBT for screening when 3D-AUS was added (P < .05 for each reader)., Conclusions: In this initial reader study, no significant difference in ROC performance was found with the addition of 3D-AUS to DBT. However, a trend to improved discrimination of malignancy was observed when adding 3D-AUS. Radiologists' confidence also improved with DBT/3DAUS compared to DBT alone.

Objective: The objective of our study was to retrospectively evaluate the imaging findings of patients with breast cancer negative for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2)-so-called "triple receptor-negative cancer"-and to compare the mammographic findings and clinical characteristics of triple receptor-negative cancer with non-triple receptor-negative cancers (i.e., ER-positive, PR-positive, or HER2-positive or two of the three markers positive)., Conclusion: Triple receptor-negative cancer was most commonly an irregular noncalcified mass with ill-defined or spiculated margins on mammography and a hypoechoic or complex mass with an irregular shape and noncircumscribed margins on ultrasound. Most triple receptor-negative cancers were discovered on physical examination. Compared with non-triple receptor-negative cancers, triple receptor-negative cancers were found in younger women and were a higher pathologic grade.

Purpose: The authors are developing a content-based image retrieval (CBIR) CADx system to assist radiologists in characterization of breast masses on ultrasound images. In this study, the authors compared seven similarity measures to be considered for the CBIR system. The similarity between the query and the retrieved masses was evaluated based on radiologists' visual similarity assessments., Methods: The CADx system retrieves masses that are similar to a query mass from a reference library based on computer-extracted features using a k-nearest neighbor (k-NN) approach. Among seven similarity measures evaluated for the CBIR system, four similarity measures including linear discriminant analysis (LDA), Bayesian neural network (BNN), cosine similarity measure (Cos), and Euclidean distance (ED) similarity measure were compared by radiologists' visual assessment. For LDA and BNN, the features of a query mass were combined first into a malignancy score and then masses with similar scores were retrieved. For Cos and ED, similar masses were retrieved based on the normalized dot product and the Euclidean distance, respectively, between two feature vectors. For the observer study, three most similar masses were retrieved for a given query mass with each method. All query-retrieved mass pairs were mixed and presented to the radiologists in random order. Three Mammography Quality Standards Act (MQSA) radiologists rated the similarity between each pair using a nine-point similarity scale (1 = very dissimilar, 9 = very similar). The accuracy of the CBIR CADx system using the different similarity measures to characterize malignant and benign masses was evaluated by ROC analysis., Results: The BNN measure used with the k-NN classifier provided slightly higher performance for classification of malignant and benign masses (A(z) values of 0.87) than those with the LDA, Cos, and ED measures (A(z) of 0.86, 0.84, and 0.81, respectively). The average similarity ratings of all radiologists for LDA, BNN, Cos, and ED were 4.71, 4.95, 5.18, and 5.32, respectively. The k-NN with the ED measures retrieved masses of significantly higher similarity (p < 0.008) than LDA and BNN., Conclusions: Similarity measures using the resemblance of individual features in the multidimensional feature space can retrieve visually more similar masses than similarity measures using the resemblance of the classifier scores. A CBIR system that can most effectively retrieve similar masses to the query may not have the best A(z).

Purpose: To develop a new texture-field orientation (TFO) method that combines a priori knowledge, local and global information for the automated identification of pectoral muscle on mammograms., Methods: The authors designed a gradient-based directional kernel (GDK) filter to enhance the linear texture structures, and a gradient-based texture analysis to extract a texture orientation image that represented the dominant texture orientation at each pixel. The texture orientation image was enhanced by a second GDK filter for ridge point extraction. The extracted ridge points were validated and the ridges that were less likely to lie on the pectoral boundary were removed automatically. A shortest-path finding method was used to generate a probability image that represented the likelihood that each remaining ridge point lay on the true pectoral boundary. Finally, the pectoral boundary was tracked by searching for the ridge points with the highest probability lying on the pectoral boundary. A data set of 130 MLO-view digitized film mammograms (DFMs) from 65 patients was used to train the TFO algorithm. An independent data set of 637 MLO-view DFMs from 562 patients was used to evaluate its performance. Another independent data set of 92 MLO-view full field digital mammograms (FFDMs) from 92 patients was used to assess the adaptability of the TFO algorithm to FFDMs. The pectoral boundary detection accuracy of the TFO method was quantified by comparison with an experienced radiologist's manually drawn pectoral boundary using three performance metrics: The percent overlap area (POA), the Hausdorff distance (Hdist), and the average distance (AvgDist)., Results: The mean and standard deviation of POA, Hdist, and AvgDist were 95.0 +/- 3.6%, 3.45 +/- 2.16 mm, and 1.12 +/- 0.82 mm, respectively. For the POA measure, 91.5%, 97.3%, and 98.9% of the computer detected pectoral muscles had POA larger than 90%, 85%, and 80%, respectively. For the distance measures, 85.4% and 98.0% of the computer detected pectoral boundaries had Hdist within 5 and 10 mm, respectively, and 99.4% of computer detected pectoral muscle boundaries had AvgDist within 5 mm from the radiologist's manually drawn boundaries., Conclusions: The pectoral muscle on DFMs can be detected accurately by the automated TFO method. The preliminary study of applying the same pectoral muscle identification algorithm to FFDMs without retraining demonstrates that the TFO method is reasonably robust against the differences in the image properties between the digitized and digital mammograms.

Purpose: Automated detection of breast boundary is one of the fundamental steps for computer-aided analysis of mammograms. In this study, the authors developed a new dynamic multiple thresholding based breast boundary (MTBB) detection method for digitized mammograms., Methods: A large data set of 716 screen-film mammograms (442 CC view and 274 MLO view) obtained from consecutive cases of an Institutional Review Board approved project were used. An experienced breast radiologist manually traced the breast boundary on each digitized image using a graphical interface to provide a reference standard. The initial breast boundary (MTBB-Initial) was obtained by dynamically adapting the threshold to the gray level range in local regions of the breast periphery. The initial breast boundary was then refined by using gradient information from horizontal and vertical Sobel filtering to obtain the final breast boundary (MTBB-Final). The accuracy of the breast boundary detection algorithm was evaluated by comparison with the reference standard using three performance metrics: The Hausdorff distance (HDist), the average minimum Euclidean distance (AMinDist), and the area overlap measure (AOM)., Results: In comparison with the authors' previously developed gradient-based breast boundary (GBB) algorithm, it was found that 68%, 85%, and 94% of images had HDist errors less than 6 pixels (4.8 mm) for GBB, MTBB-Initial, and MTBB-Final, respectively. 89%, 90%, and 96% of images had AMinDist errors less than 1.5 pixels (1.2 mm) for GBB, MTBB-Initial, and MTBB-Final, respectively. 96%, 98%, and 99% of images had AOM values larger than 0.9 for GBB, MTBB-Initial, and MTBB-Final, respectively. The improvement by the MTBB-Final method was statistically significant for all the evaluation measures by the Wilcoxon signed rank test (p < 0.0001)., Conclusions: The MTBB approach that combined dynamic multiple thresholding and gradient information provided better performance than the breast boundary detection algorithm that mainly used gradient information.

The goal of this study was to develop an automated method to segment breast masses on dynamic contrast-enhanced (DCE) magnetic resonance (MR) scans and to evaluate its potential for estimating tumor volume on pre- and postchemotherapy images and tumor change in response to treatment. A radiologist experienced in interpreting breast MR scans defined a cuboid volume of interest (VOI) enclosing the mass in the MR volume at one time point within the sequence of DCE-MR scans. The corresponding VOIs over the entire time sequence were then automatically extracted. A new 3D VOI representing the local pharmacokinetic activities in the VOI was generated from the 4D VOI sequence by summarizing the temporal intensity enhancement curve of each voxel with its standard deviation. The method then used the fuzzy c-means (FCM) clustering algorithm followed by morphological filtering for initial mass segmentation. The initial segmentation was refined by the 3D level set (LS) method. The velocity field of the LS method was formulated in terms of the mean curvature which guaranteed the smoothness of the surface, the Sobel edge information which attracted the zero LS to the desired mass margin, and the FCM membership function which improved segmentation accuracy. The method was evaluated on 50 DCE-MR scans of 25 patients who underwent neoadjuvant chemotherapy. Each patient had pre- and postchemotherapy DCE-MR scans on a 1.5 T magnet. The in-plane pixel size ranged from 0.546 to 0.703 mm and the slice thickness ranged from 2.5 to 4.5 mm. The flip angle was 15 degrees, repetition time ranged from 5.98 to 6.7 ms, and echo time ranged from 1.2 to 1.3 ms. Computer segmentation was applied to the coronal T1-weighted images. For comparison, the same radiologist who marked the VOI also manually segmented the mass on each slice. The performance of the automated method was quantified using an overlap measure, defined as the ratio of the intersection of the computer and the manual segmentation volumes to the manual segmentation volume. Pre- and postchemotherapy masses had overlap measures of 0.81 +/- 0.13 (mean +/- s.d.) and 0.71 +/- 0.22, respectively. The percentage volume reduction (PVR) estimated by computer and the radiologist were 55.5 +/- 43.0% (mean +/- s.d.) and 57.8 +/- 51.3%, respectively. Paired Student's t test indicated that the difference between the mean PVRs estimated by computer and the radiologist did not reach statistical significance (p = 0.641). The automated mass segmentation method may have the potential to assist physicians in monitoring volume change in breast masses in response to treatment.

Rationale and Objectives: To investigate the effect of a computer-aided diagnosis (CADx) system on radiologists' performance in discriminating malignant and benign masses on mammograms and three-dimensional (3D) ultrasound (US) images., Materials and Methods: Our dataset contained mammograms and 3D US volumes from 67 women (median age, 51; range: 27-86) with 67 biopsy-proven breast masses (32 benign and 35 malignant). A CADx system was designed to automatically delineate the mass boundaries on mammograms and the US volumes, extract features, and merge the extracted features into a multi-modality malignancy score. Ten experienced readers (subspecialty academic breast imaging radiologists) first viewed the mammograms alone, and provided likelihood of malignancy (LM) ratings and Breast Imaging and Reporting System assessments. Subsequently, the reader viewed the US images with the mammograms, and provided LM and action category ratings. Finally, the CADx score was shown and the reader had the opportunity to revise the ratings. The LM ratings were analyzed using receiver-operating characteristic (ROC) methodology, and the action category ratings were used to determine the sensitivity and specificity of cancer diagnosis., Results: Without CADx, readers' average area under the ROC curve, A(z), was 0.93 (range, 0.86-0.96) for combined assessment of the mass on both the US volume and mammograms. With CADx, their average A(z) increased to 0.95 (range, 0.91-0.98), which was borderline significant (P = .05). The average sensitivity of the readers increased from 98% to 99% with CADx, while the average specificity increased from 27% to 29%. The change in sensitivity with CADx did not achieve statistical significance for the individual radiologists, and the change in specificity was statistically significant for one of the radiologists., Conclusions: A well-trained CADx system that combines features extracted from mammograms and US images may have the potential to improve radiologists' performance in distinguishing malignant from benign breast masses and making decisions about biopsies.

Segmentation is one of the first steps in most computer-aided diagnosis systems for characterization of masses as malignant or benign. In this study, the authors designed an automated method for segmentation of breast masses on ultrasound (US) images. The method automatically estimated an initial contour based on a manually identified point approximately at the mass center. A two-stage active contour method iteratively refined the initial contour and performed self-examination and correction on the segmentation result. To evaluate the method, the authors compared it with manual segmentation by two experienced radiologists (R1 and R2) on a data set of 488 US images from 250 biopsy-proven masses (100 malignant and 150 benign). Two area overlap ratios (AOR1 and AOR2) and an area error measure were used as performance measures to evaluate the segmentation accuracy. Values for AOR1, defined as the ratio of the intersection of the computer and the reference segmented areas to the reference segmented area, were 0.82 +/- 0.16 and 0.84 +/- 0.18, respectively, when manually segmented mass regions by R1 and R2 were used as the reference. Although this indicated a high agreement between the computer and manual segmentations, the two radiologists' manual segmentation results were significantly (p < 0.03) more consistent, with AOR1 = 0.84 +/- 0.16 and 0.91 +/- 0.12, respectively, when the segmented regions by R1 and R2 were used as the reference. To evaluate the segmentation method in terms of lesion classification accuracy, feature spaces were formed by extracting texture, width-to-height, and posterior shadowing features based on either automated computer segmentation or the radiologists' manual segmentation. A linear discriminant analysis classifier was designed using stepwise feature selection and two-fold cross validation to characterize the mass as malignant or benign. For features extracted from computer segmentation, the case-based test A(z) values ranged from 0.88 +/- 0.03 to 0.92 +/- 0.02, indicating a comparable performance to those extracted from manual segmentation by radiologists (A(z) value range: 0.87 +/- 0.03 to 0.90 +/- 0.03).

Purpose: To assess the diagnostic performance of various Doppler ultrasonographic (US) vascularity measures in conjunction with grayscale (GS) criteria in differentiating benign from malignant breast masses, by using histologic findings as the reference standard., Materials and Methods: Institutional Review Board and HIPAA standards were followed. Seventy-eight women (average age, 49 years; range, 26-70 years) scheduled for breast biopsy were included. Thirty-eight patient scans were partially analyzed and published previously, and 40 additional scans were used as a test set to evaluate previously determined classification indexes. In each patient, a series of color Doppler images was acquired and reconstructed into a volume encompassing a suspicious mass, identified by a radiologist-defined ellipsoid, in which six Doppler vascularity measures were calculated. Radiologist GS ratings and patient age were also recorded. Multivariable discrimination indexes derived from the learning set were applied blindly to the test set. Overall performance was also confirmed by using a fourfold cross-validation scheme on the entire population., Results: By using all cases (46 benign, 32 malignant), the area under the receiver operating characteristic curve (A(z)) values confirmed results of previous analyses: Speed-weighted pixel density (SWPD) performed the best as a diagnostic index, although statistical significance (P = .01) was demonstrated only with respect to the normalized power-weighted pixel density. In both learning and test sets, the three-variable index (SWPD-age-GS) displayed significantly better diagnostic performance (A(z) = 0.97) than did any single index or the one two-variable index (age-GS) that could be obtained without the data from the Doppler scan. Results of the cross validation confirmed the trends in the two data sets., Conclusion: Quantitative Doppler US vascularity measurements considerably contribute to malignant breast tissue identification beyond subjective GS evaluation alone. The SWPD-age-GS index has high performance (A(z) = 0.97), regardless of incidental performance variations in its single variable components., ((c) RSNA, 2008.)

Rationale and Objectives: To compare the performance of computer aided detection (CAD) systems on pairs of full-field digital mammogram (FFDM) and screen-film mammogram (SFM) obtained from the same patients., Materials and Methods: Our CAD systems on both modalities have similar architectures that consist of five steps. For FFDMs, the input raw image is first log-transformed and enhanced by a multiresolution preprocessing scheme. For digitized SFMs, the input image is smoothed and subsampled to a pixel size of 100 microm x 100 microm. For both CAD systems, the mammogram after preprocessing undergoes a gradient field analysis followed by clustering-based region growing to identify suspicious breast structures. Each of these structures is refined in a local segmentation process. Morphologic and texture features are then extracted from each detected structure, and trained rule-based and linear discriminant analysis classifiers are used to differentiate masses from normal tissues. Two datasets, one with masses and the other without masses, were collected. The mass dataset contained 131 cases with 131 biopsy proven masses, of which 27 were malignant and 104 benign. The true locations of the masses were identified by an experienced Mammography Quality Standards Act (MQSA) radiologist. The no-mass data set contained 98 cases. The time interval between the FFDM and the corresponding SFM was 0 to 118 days., Results: Our CAD system achieved case-based sensitivities of 70%, 80%, and 90% at 0.9, 1.5, and 2.6 false positive (FP) marks/image, respectively, on FFDMs, and the same sensitivities at 1.0, 1.4, and 2.6 FP marks/image, respectively, on SFMs., Conclusions: The difference in the performances of our FFDM and SFM CAD systems did not achieve statistical significance.

Purpose: To retrospectively investigate the effect of using a custom-designed computer classifier on radiologists' sensitivity and specificity for discriminating malignant masses from benign masses on three-dimensional (3D) volumetric ultrasonographic (US) images, with histologic analysis serving as the reference standard., Materials and Methods: Informed consent and institutional review board approval were obtained. Our data set contained 3D US volumetric images obtained in 101 women (average age, 51 years; age range, 25-86 years) with 101 biopsy-proved breast masses (45 benign, 56 malignant). A computer algorithm was designed to automatically delineate mass boundaries and extract features on the basis of segmented mass shapes and margins. A computer classifier was used to merge features into a malignancy score. Five experienced radiologists participated as readers. Each radiologist read cases first without computer-aided diagnosis (CAD) and immediately thereafter with CAD. Observers' malignancy rating data were analyzed with the receiver operating characteristic (ROC) curve., Results: Without CAD, the five radiologists had an average area under the ROC curve (A(z)) of 0.83 (range, 0.81-0.87). With CAD, the average A(z) increased significantly (P = .006) to 0.90 (range, 0.86-0.93). When a 2% likelihood of malignancy was used as the threshold for biopsy recommendation, the average sensitivity of radiologists increased from 96% to 98% with CAD, while the average specificity for this data set decreased from 22% to 19%. If a biopsy recommendation threshold could be chosen such that sensitivity would be maintained at 96%, specificity would increase to 45% with CAD., Conclusion: Use of a computer algorithm may improve radiologists' accuracy in distinguishing malignant from benign breast masses on 3D US volumetric images., ((c) RSNA, 2007.)

Purpose: To retrospectively compare computer-aided mammographic density estimation (MDEST) with radiologist estimates of percentage density and Breast Imaging Reporting and Data System (BI-RADS) density classification., Materials and Methods: Institutional Review Board approval was obtained for this HIPAA-compliant study; patient informed consent requirements were waived. A fully automated MDEST computer program was used to measure breast density on digitized mammograms in 65 women (mean age, 53 years; range, 24-89 years). Pixel gray levels in detected breast borders were analyzed, and dense areas were segmented. Percentage density was calculated by dividing the number of dense pixels by the total number of pixels within the borders. Seven breast radiologists (five trained with MDEST, two not trained) prospectively assigned qualitative BI-RADS density categories and visually estimated percentage density on 260 mammograms. Qualitative BI-RADS assessments were compared with new quantitative BI-RADS standards. The reference standard density for this study was established by allowing the five trained radiologists to manipulate the MDEST gray-level thresholds, which segmented mammograms into dense and nondense areas. Statistical tests performed include Pearson correlation coefficients, Bland-Altman agreement method, kappa statistics, and unpaired t tests., Results: There was a close correlation between the reference standard and radiologist-estimated density (R = 0.90-0.95) and MDEST density (R = 0.89). Untrained radiologists overestimated percentage density by an average of 37%, versus 6% for trained radiologists (P < .001). MDEST showed better agreement with the reference standard (average overestimate, 1%; range, -15% to +18%). MDEST correlated better with percentage density than with qualitative BI-RADS categories. There were large overlaps and ranges of percentage density in qualitative BI-RADS categories 2-4. Qualitative BI-RADS categories correlated poorly with new quantitative BI-RADS categories, and 16 (6%) of 260 views were erroneously classified by MDEST., Conclusion: MDEST compared favorably with radiologist estimates of percentage density and is more reproducible than radiologist estimates when qualitative BI-RADS density categories are used. Qualitative and quantitative BI-RADS density assessments differed markedly., ((c) RSNA, 2006.)

Purpose: To retrospectively evaluate effects of computer-aided diagnosis (CAD) involving an interval change classifier (which uses interval change information extracted from prior and current mammograms and estimates a malignancy rating) on radiologists' accuracy in characterizing masses on two-view serial mammograms as malignant or benign., Materials and Methods: The data collection protocol had institutional review board approval. Patient informed consent was waived for this HIPAA-compliant retrospective study. Ninety temporal pairs of two-view serial mammograms (depicting 47 malignant and 43 benign biopsy-proved masses) were obtained from 68 patient files and were digitized. Biopsy was the reference standard. Eight Mammography Quality Standards Act of 1992-accredited radiologists and two breast imaging fellows assessed digitized two-view temporal pairs (in preselected regions of interest only) by estimating likelihood of malignancy and Breast Imaging Reporting and Data System (BI-RADS) category without and with CAD. Observers' rating data were analyzed with Dorfman-Berbaum-Metz (DBM) multireader multicase method. Statistical significance of differences was estimated with the DBM method and Student two-tailed paired t test., Results: Average area under the receiver operating characteristic curve for likelihood of malignancy across the 10 observers was 0.83 (range, 0.74-0.88) without CAD and improved to 0.87 (range, 0.80-0.92) with CAD (P < .05). The average partial area index above a sensitivity of 0.90 for likelihood of malignancy was 0.35 (range, 0.13-0.54) without CAD and 0.49 (range, 0.18-0.73) with CAD--a nonsignificant improvement (P = .11). For BI-RADS assessment, it was estimated that with CAD, six radiologists would correctly recommend additional biopsies for malignant masses (range, 4.3%-10.6%) and five would correctly recommend reduction of biopsy (ie, fewer biopsies) for benign masses (range, 2.3%-9.3%). However, five radiologists would incorrectly recommend additional biopsy for benign masses (range, 2.3%-14.0%), and one would incorrectly recommend reduction of biopsy (4.3%)., Conclusion: CAD involving interval change analysis of preselected regions of interest can significantly improve radiologists' accuracy in classifying masses on digitized screen-film mammograms as malignant or benign., (RSNA, 2006)

We are developing new techniques to improve the accuracy of computerized microcalcification detection by using the joint two-view information on craniocaudal (CC) and mediolateral-oblique (MLO) views. After cluster candidates were detected using a single-view detection technique, candidates on CC and MLO views were paired using their radial distances from the nipple. Candidate pairs were classified with a similarity classifier that used the joint information from both views. Each cluster candidate was also characterized by its single-view features. The outputs of the similarity classifier and the single-view classifier were fused and the cluster candidate was classified as a true microcalcification cluster or a false-positive (FP) using the fused two-view information. A data set of 116 pairs of mammograms containing microcalcification clusters and 203 pairs of normal images from the University of South Florida (USF) public database was used for training the two-view detection algorithm. The trained method was tested on an independent test set of 167 pairs of mammograms, which contained 71 normal pairs and 96 pairs with microcalcification clusters collected at the University of Michigan (UM). The similarity classifier had a very low FP rate for the test set at low and medium levels of sensitivity. However, the highest mammogram-based sensitivity that could be reached by the similarity classifier was 69%. The single-view classifier had a higher FP rate compared to the similarity classifier, but it could reach a maximum mammogram-based sensitivity of 93%. The fusion method combined the scores of these two classifiers so that the number of FPs was substantially reduced at relatively low and medium sensitivities, and a relatively high maximum sensitivity was maintained. For the malignant microcalcification clusters, at a mammogram-based sensitivity of 80%, the FP rates were 0.18 and 0.35 with the two-view fusion and single-view detection methods, respectively. When the training and test sets were switched, a similar improvement was obtained, except that both the fusion and single-view detection methods had superior test performances on the USF data set than those on the UM data set. Our results indicate that correspondence of cluster candidates on two different views provides valuable additional information for distinguishing FPs from true microcalcification clusters.

Background: The goals of the current study were to compare the clinicopathologic presentations of patients with lobular carcinoma in situ (LCIS) as a component of breast carcinoma who were treated with breast conserving surgery (BCS) and radiation therapy (RT) with those of patients without LCIS as part of their primary tumor and to report rates of local control by overall cohort and specifically in patients with positive margins for LCIS and multifocal LCIS., Methods: Sixty-four patients with Stages 0-II breast carcinoma with LCIS (LCIS-containing tumor group, LCTG) that had received BCS+RT treatment at the University of Michigan between 1989 and 2003 were identified. These patients were matched to 121 patients without LCIS (control group) in a 1:2 ratio., Results: The median follow-up time was 3.9 years (range, 0.3-18.9 yrs). There were no significant differences between the two groups with regard to clinical, pathologic, or treatment-related variables or in mammographic presentation, with the exception of a higher proportion of the LCTG patients who received adjuvant hormonal therapy (P = 0.01). The rates of local control at 5 years were 100% in the LCTG group and 99.1% in the control group (P = 0.86). The presence of LCIS at the margins and the size and presence of multifocal LCIS did not alter the rate of local control., Conclusions: The extent of LCIS and its presence at the margins did not reduce the excellent rates of local control after BCS+RT. The data suggest that LCIS in the tumor specimen, even when multifocal, should not affect selection of patients for BCS and whole-breast RT., (Copyright 2005 American Cancer Society.)

Purpose: To prospectively compare the ability of clinical examination, mammography, vascularity-sensitive ultrasound, and magnetic resonance imaging (MRI) to determine pathologic complete response (CR) in breast cancer patients undergoing neoadjuvant chemotherapy., Patients and Methods: Participants were women with primary measurable, operable invasive breast cancer (Stages I-III) who presented to the University of Michigan Breast Care Center. Eligibility criteria were based on clinical need for chemotherapy as part of the overall treatment plan. The chemotherapy consisted of doxorubicin and docetaxel administered every 3 weeks for four cycles. Tumor size measurements by physical examination and by the three imaging modalities were performed before chemotherapy was initiated and after its completion, prior to definitive surgery. Response criteria were pre-specified in this prospective design, and study radiologists analyzed the mammographic, sonographic and MRI image sets blinded to information from the other modalities and blinded to final histological diagnosis. The pathologic CR rate obtained by the clinical and imaging modalities was compared to pathologic CR as determined pathologically., Results: 41 of 43 enrolled patients had a determination of pathologic response, and 4 patients had a pathologic CR to this chemotherapy (9.8%). The accuracy of physical examination, mammography, ultrasound, and MRI in determining pathologic CR was 75, 89, 82, and 89% respectively (NS)., Conclusion: Biopsy after neoadjuvant chemotherapy remains absolutely necessary to determine pathologic CR to neoadjuvant chemotherapy, as the accuracy of current imaging modalities is insufficient to make this determination. The accuracy of mammography, vascularity-sensitive ultrasound, and MRI were not observed to be significantly different.

Purpose: To evaluate the effects of computer-aided diagnosis (CAD) on radiologists' characterization of masses on serial mammograms., Materials and Methods: Two hundred fifty-three temporal image pairs (138 malignant and 115 benign) obtained from 96 patients who had masses on serial mammograms were evaluated. The temporal pairs were formed by matching masses of the same view from two different examinations. Eight radiologists and two breast imaging fellows assessed the temporal pairs with and without computer aid. The classification of accuracy was quantified by using the area under receiver operating characteristic curve (A(z)). The statistical significance of the difference in A(z) between the different reading conditions was estimated with the Dorfman-Berbaum-Metz method for analysis of multireader multicase data and with the Student paired t test for analysis of observer-specific paired data., Results: The average A(z) for radiologists' estimates of the likelihood of malignancy was 0.79 without CAD and improved to 0.84 with CAD. The improvement was statistically significant (P =.005). The corresponding average partial area index was 0.25 without CAD and improved to 0.37 with CAD. The improvement was also statistically significant (P =.005). On the basis of Breast Imaging Reporting and Data System assessments, it was estimated that with CAD, each radiologist, on average, reduced 0.7% (0.8 of 115) of unnecessary biopsies and correctly recommended 5.7% (7.8 of 138) of additional biopsies., Conclusion: CAD based on analysis of interval changes can significantly increase radiologists' accuracy in classification of masses and thereby may be useful in improving correct biopsy recommendations., (Copyright RSNA, 2004)

Correlation of information from multiple-view mammograms (e.g., MLO and CC views, bilateral views, or current and prior mammograms) can improve the performance of breast cancer diagnosis by radiologists or by computer. The nipple is a reliable and stable landmark on mammograms for the registration of multiple mammograms. However, accurate identification of nipple location on mammograms is challenging because of the variations in image quality and in the nipple projections, resulting in some nipples being nearly invisible on the mammograms. In this study, we developed a computerized method to automatically identify the nipple location on digitized mammograms. First, the breast boundary was obtained using a gradient-based boundary tracking algorithm, and then the gray level profiles along the inside and outside of the boundary were identified. A geometric convergence analysis was used to limit the nipple search to a region of the breast boundary. A two-stage nipple detection method was developed to identify the nipple location using the gray level information around the nipple, the geometric characteristics of nipple shapes, and the texture features of glandular tissue or ducts which converge toward the nipple. At the first stage, a rule-based method was designed to identify the nipple location by detecting significant changes of intensity along the gray level profiles inside and outside the breast boundary and the changes in the boundary direction. At the second stage, a texture orientation-field analysis was developed to estimate the nipple location based on the convergence of the texture pattern of glandular tissue or ducts towards the nipple. The nipple location was finally determined from the detected nipple candidates by a rule-based confidence analysis. In this study, 377 and 367 randomly selected digitized mammograms were used for training and testing the nipple detection algorithm, respectively. Two experienced radiologists identified the nipple locations which were used as the gold standard. In the training data set, 301 nipples were positively identified and were referred to as visible nipples. Seventy six nipples could not be positively identified and were referred to as invisible nipples. The radiologists provided their estimation of the nipple locations in the latter group for comparison with the computer estimates. The computerized method could detect 89.37% (269/301) of the visible nipples and 69.74% (53/76) of the invisible nipples within 1 cm of the gold standard. In the test data set, 298 and 69 of the nipples were classified as visible and invisible, respectively. 92.28% (275/298) of the visible nipples and 53.62% (37/69) of the invisible nipples were identified within 1 cm of the gold standard. The results demonstrate that the nipple locations on digitized mammograms can be accurately detected if they are visible and can be reasonably estimated if they are invisible. Automated nipple detection will be an important step towards multiple image analysis for CAD.

Purpose: To retrospectively determine frequency of invasive cancer or ductal carcinoma in situ (DCIS) at excisional biopsy in women with atypical lobular hyperplasia (ALH) or lobular carcinoma in situ (LCIS) at percutaneous core-needle biopsy (CNB)., Materials and Methods: Review of results in 6,081 consecutive patients who underwent CNB at two institutions revealed that in 35 (0.58%), LCIS (n = 15) or ALH (n = 20) was the pathologic finding with highest risk. Patient age range was 41-84 years (mean, 59 years). Of 35 patients, 26 (74%) underwent excisional biopsy and nine (26%) underwent mammographic follow-up for longer than 2 years. Lesions with a pathologic upgrade were noted when invasive cancer or DCIS occurred at the CNB site. CNB results in patients with a diagnosis of atypical ductal hyperplasia (ADH) (75 of 6,081 [1.2%]) were reviewed; these patients underwent subsequent excisional biopsy. Statistical comparison of frequency of upgrading of lesions in patients with a diagnosis of LCIS or ALH at CNB and in those with a diagnosis of ADH at CNB was performed (Pearson chi(2) test)., Results: In six (17%) of 35 (95% CI: 4.7%, 29.6%) patients, lesions were upgraded to DCIS (n = 4) or invasive cancer (n = 2). In 15 patients with LCIS diagnosed at CNB, lesions in four (27%) were upgraded to either DCIS or invasive cancer. In 20 patients with ALH diagnosed at CNB, lesions were upgraded to DCIS in two (10%). Lesions in nine patients who underwent mammographic follow-up were stable. No mammographic or technical findings distinguished patients with upgraded lesions from those whose lesions were not upgraded. In 12 (16%) of 75 (95% CI: 7.7%, 24.3%) patients with ADH, lesions were upgraded. Difference between the upgrade rate in patients with LCIS or ALH and that in those with ADH was not significant (P =.88)., Conclusion: Lesions in 17% of patients with LCIS or ALH at CNB were upgraded to invasive cancer or DCIS; this rate was similar to the upgrade rate in patients with ADH. Excisional biopsy is supported when LCIS, ALH, or ADH is diagnosed at CNB., (Copyright RSNA, 2004)

We are developing an automated stereo spot mammography technique for improved imaging of suspicious dense regions within digital mammograms. The technique entails the acquisition of a full-field digital mammogram, automated detection of a suspicious dense region within that mammogram by a computer aided detection (CAD) program, and acquisition of a stereo pair of images with automated collimation to the suspicious region. The latter stereo spot image is obtained within seconds of the original full-field mammogram, without releasing the compression paddle. The spot image is viewed on a stereo video display. A critical element of this technique is the automated detection of suspicious regions for spot imaging. We performed an observer study to compare the suspicious regions selected by radiologists with those selected by a CAD program developed at the University of Michigan. True regions of interest (TROIs) were separately determined by one of the radiologists who reviewed the original mammograms, biopsy images, and histology results. We compared the radiologist and computer-selected regions of interest (ROIs) to the TROIs. Both the radiologists and the computer were allowed to select up to 3 regions in each of 200 images (mixture of 100 CC and 100 MLO views). We computed overlap indices (the overlap index is defined as the ratio of the area of intersection to the area of interest) to quantify the agreement between the selected regions in each image. The averages of the largest overlap indices per image for the 5 radiologist-to-computer comparisons were directly related to the average number of regions per image traced by the radiologists (about 50% for 1 region/image, 84% for 2 regions/image and 96% for 3 regions/image). The average of the overlap indices with all of the TROIs was 73% for CAD and 76.8% +/- 10.0% for the radiologists. This study indicates that the CAD determined ROIs could potentially be useful for a screening technique that includes stereo spot mammography imaging.

Purpose: To evaluate a noncommercial computer-aided detection (CAD) program for breast cancer detection with screening mammography., Materials and Methods: A CAD program was developed for mammographic breast cancer detection. The program was applied to 2,389 patients' screening mammograms at two geographically remote academic institutions (institutions A and B). Thirteen radiologists who specialized in breast imaging participated in this pilot study. For each case, the individual radiologist performed a prospective Breast Imaging Reporting and Data System (BI-RADS) assessment after viewing of the screening mammogram. Subsequently, the radiologist was shown CAD results and rendered a second BI-RADS assessment by using knowledge of both mammographic appearance and CAD results. Outcome analysis of results of examination in patients recalled for a repeat examination, of biopsy, and of 1-year follow-up examination was recorded. Correct detection with CAD included a computer-generated mark indicating a possible malignancy on craniocaudal or mediolateral oblique views or both., Results: Eleven (0.46%) of 2,389 patients had mammographically detected nonpalpable breast cancers. Ten (91%) of 11 (95% CI: 74%, 100%) cancers were correctly identified with CAD. Radiologist sensitivity without CAD was 91% (10 of 11; 95% CI: 74%, 100%). In 1,077 patients, follow-up findings were documented at 1 year. Five (0.46%) patients developed cancers, which were found on subsequent screening mammograms. The area where the cancers developed in two (40%) of these five patients was marked (true-positive finding) by the computer in the preceding year. Because of CAD results, a 9.7% increase in recall rate from 14.4% (344 of 2,389) to 15.8% (378 of 2,389) occurred. Radiologists' recall rate of study patients prior to use of CAD was 31% higher than the average rate for nonstudy cases (10.3%) during the same time period at institution A., Conclusion: Performance of the CAD program had a very high sensitivity of 91% (95% CI: 74%, 100%)., (Copyright RSNA, 2004)

We are developing computer vision techniques for the characterization of breast masses as malignant or benign on radiologic examinations. In this study, we investigated the computerized characterization of breast masses on three-dimensional (3-D) ultrasound (US) volumetric images. We developed 2-D and 3-D active contour models for automated segmentation of the mass volumes. The effect of the initialization method of the active contour on the robustness of the iterative segmentation method was studied by varying the contour used for its initialization. For a given segmentation, texture and morphological features were automatically extracted from the segmented masses and their margins. Stepwise discriminant analysis with the leave-one-out method was used to select effective features for the classification task and to combine these features into a malignancy score. The classification accuracy was evaluated using the area Az under the receiver operating characteristic (ROC) curve, as well as the partial area index Az(0.9), defined as the relative area under the ROC curve above a sensitivity threshold of 0.9. For the purpose of comparison with the computer classifier, four experienced breast radiologists provided malignancy ratings for the 3-D US masses. Our dataset consisted of 3-D US volumes of 102 biopsied masses (46 benign, 56 malignant). The classifiers based on 2-D and 3-D segmentation methods achieved test Az values of 0.87+/-0.03 and 0.92+/-0.03, respectively. The difference in the Az values of the two computer classifiers did not achieve statistical significance. The Az values of the four radiologists ranged between 0.84 and 0.92. The difference between the computer's Az value and that of any of the four radiologists did not achieve statistical significance either. However, the computer's Az(0.9) value was significantly higher than that of three of the four radiologists. Our results indicate that an automated and effective computer classifier can be designed for differentiating malignant and benign breast masses on 3-D US volumes. The accuracy of the classifier designed in this study was similar to that of experienced breast radiologists.

Mass segmentation is used as the first step in many computer-aided diagnosis (CAD) systems for classification of breast masses as malignant or benign. The goal of this paper was to study the accuracy of an automated mass segmentation method developed in our laboratory, and to investigate the effect of the segmentation stage on the overall classification accuracy. The automated segmentation method was quantitatively compared with manual segmentation by two expert radiologists (R1 and R2) using three similarity or distance measures on a data set of 100 masses. The area overlap measures between R1 and R2, the computer and R1, and the computer and R2 were 0.76 +/- 0.13, 0.74 +/- 0.11, and 0.74 +/- 0.13, respectively. The interobserver difference in these measures between the two radiologists was compared with the corresponding differences between the computer and the radiologists. Using three similarity measures and data from two radiologists, a total of six statistical tests were performed. The difference between the computer and the radiologist segmentation was significantly larger than the interobserver variability in only one test. Two sets of texture, morphological, and spiculation features, one based on the computer segmentation, and the other based on radiologist segmentation, were extracted from a data set of 249 films from 102 patients. A classifier based on stepwise feature selection and linear discriminant analysis was trained and tested using the two feature sets. The leave-one-case-out method was used for data sampling. For case-based classification, the area Az under the receiver operating characteristic (ROC) curve was 0.89 and 0.88 for the feature sets based on the radiologist segmentation and computer segmentation, respectively. The difference between the two ROC curves was not statistically significant.

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.

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.

Objective: Patients who have had cancer in one breast are at high risk for cancer in the contralateral breast. These bilateral cancers may be synchronous or metachronous. If the manifestations on mammography were similar in both breasts, an aggressive search for the mammographic findings of the first breast cancer might lead to early detection of the contralateral cancer. The purpose of this study was to evaluate mammograms for patients with bilateral cancers to determine whether the mammographic appearance of the contralateral cancer is likely to be the same as that of the first cancer., Materials and Methods: We retrospectively reviewed the pathologic and mammographic records of 69 patients with surgically proven bilateral primary breast cancer. Thirty four of 69 (49%) had synchronous cancer, and 35 (51%) had metachronous cancer. Mammographic appearances were classified as microcalcifications, spiculated mass, nonspiculated mass (whether circumscribed or poorly defined), asymmetric or developing density, architectural distortion, and normal. Multiple findings were subclassified as major and minor findings. All findings were compared between both breast cancers, and statistical significance was determined by the two-sample Z test., Results: Forty six (67%) of 69 patients had different major mammographic findings in the contralateral cancer. Of 30 patients whose first cancers had microcalcifications, 20 (67%) had microcalcifications in the contralateral cancer. Of 39 patients whose first cancers lacked microcalcifications, 17 (44%) had microcalcifications in the contralateral cancer. This difference was statistically significant (p = .02). Of 26 patients whose first cancers had spiculated masses, 9 (35%) had a contralateral spiculated mass. Of 43 patients whose first cancers lacked spiculated masses, 12 (28%) had a contralateral spiculated mass. This difference was not statistically significant (p = .22)., Conclusion: Our results show that contralateral tumors usually have major mammographic findings different from those of the first cancer, and the mammographic signs of the first cancer do not indicate the most likely appearance of cancer in the contralateral breast. Evaluation of a contralateral mammogram should be performed without regard for the mammographic findings for the first cancer.

Rationale and Objectives: Mammographic findings and method of detection of 52 cases of invasive lobular carcinoma (ILC), the second most common breast carcinoma, are reported., Methods: Preoperative mammograms and clinical records of all patients with ILC not associated with a second mammary carcinoma (other than lobular carcinoma in situ) from 1979-1991 at the authors' institution were retrospectively reviewed., Results: Abnormal mammographic findings were present in 48/52 (92%) and included irregular spiculated masses (33/52, 63%), asymmetric densities (7/52, 13%), architectural distortion (5/52, 10%), microcalcifications (2/52, 4%), and well circumscribed masses (1/52, 2%). The mean mammographic diameter was 2.1 cm. The tumor was most often best visualized in the craniocaudal projection. At the time of diagnosis, 54% of women had coexistent suggestive breast physical findings and 35% had metastatic carcinoma in axillary lymph nodes., Conclusions: The infrequency of microcalcifications in pure ILC may hinder mammographic detection and contrasts markedly with ductal carcinoma. Mammography and breast physical examination play complementary roles in the detection of ILC.

Computed tomography (CT), xeroradiography, and radiography were compared in vitro to assess the relative value of each in detecting soft-tissue foreign bodies. Results indicate that CT may prove useful.