Computer-aided detection system for breast masses on digital tomosynthesis mammograms: preliminary experience

Typical values of z-resolution for different Digital Breast Tomosynthesis systems evaluated in a multicenter study

PURPOSE

The aim of the present study, conducted by a working group of the Italian Association of Medical Physics (AIFM), was to define typical z-resolution values for different digital breast tomosynthesis (DBT) models to be used as a reference for quality control (QC). Currently, there are no typical values published in internationally agreed QC protocols.

METHODS

To characterize the z-resolution of the DBT models, the full width at half maximum (FWHM) of the artifact spread function (ASF), a technical parameter that quantifies the signal intensity of a detail along reconstructed planes, was analyzed. Five different commercial phantoms, CIRS Model 011, CIRS Model 015, Modular DBT phantom, Pixmam 3-D, and Tomophan, were evaluated on reconstructed DBT images and 82 DBT systems (6 vendors, 9 models) in use at 39 centers in Italy were involved.

RESULTS

The ASF was found to be dependent on the detail size, the DBT angular acquisition range, the reconstruction algorithm and applied image processing. In particular, a progressively greater signal spread was observed as the detail size increased and the acquisition angle decreased. However, a clear correlation between signal spread and angular range width was not observed due to the different signal reconstruction and image processing strategies implemented in the algorithms developed by the vendors studied.

CONCLUSIONS

The analysis led to the identification of typical z-resolution values for different DBT model-phantom configurations that could be used as a reference during a QC program.

AI-based computer-aided diagnostic system of chest digital tomography synthesis: Demonstrating comparative advantage with X-ray-based AI systems

BACKGROUND

Compared with chest X-ray (CXR) imaging, which is a single image projected from the front of the patient, chest digital tomosynthesis (CDTS) imaging can be more advantageous for lung lesion detection because it acquires multiple images projected from multiple angles of the patient. Various clinical comparative analysis and verification studies have been reported to demonstrate this, but there is no artificial intelligence (AI)-based comparative analysis studies. Existing AI-based computer-aided detection (CAD) systems for lung lesion diagnosis have been developed mainly based on CXR images; however, CAD-based on CDTS, which uses multi-angle images of patients in various directions, has not been proposed and verified for its usefulness compared to CXR-based counterparts.

BACKGROUND AND OBJECTIVE

This study develops and tests a CDTS-based AI CAD system to detect lung lesions to demonstrate performance improvements compared to CXR-based AI CAD.

METHODS

We used multiple (e.g., five) projection images as input for the CDTS-based AI model and a single-projection image as input for the CXR-based AI model to compare and evaluate the performance between models. Multiple/single projection input images were obtained by virtual projection on the three-dimensional (3D) stack of computed tomography (CT) slices of each patient's lungs from which the bed area was removed. These multiple images result from shooting from the front and left and right 30/60∘. The projected image captured from the front was used as the input for the CXR-based AI model. The CDTS-based AI model used all five projected images. The proposed CDTS-based AI model consisted of five AI models that received images in each of the five directions, and obtained the final prediction result through an ensemble of five models. Each model used WideResNet-50. To train and evaluate CXR- and CDTS-based AI models, 500 healthy data, 206 tuberculosis data, and 242 pneumonia data were used, and three three-fold cross-validation was applied.

RESULTS

The proposed CDTS-based AI CAD system yielded sensitivities of 0.782 and 0.785 and accuracies of 0.895 and 0.837 for the (binary classification) performance of detecting tuberculosis and pneumonia, respectively, against normal subjects. These results show higher performance than the sensitivity of 0.728 and 0.698 and accuracies of 0.874 and 0.826 for detecting tuberculosis and pneumonia through the CXR-based AI CAD, which only uses a single projection image in the frontal direction. We found that CDTS-based AI CAD improved the sensitivity of tuberculosis and pneumonia by 5.4% and 8.7% respectively, compared to CXR-based AI CAD without loss of accuracy.

CONCLUSIONS

This study comparatively proves that CDTS-based AI CAD technology can improve performance more than CXR. These results suggest that we can enhance the clinical application of CDTS. Our code is available at https://github.com/kskim-phd/CDTS-CAD-P.

Artificial intelligence for digital breast tomosynthesis: Impact on diagnostic performance, reading times, and workload in the era of personalized screening

The ultimate goals of the application of artificial intelligence (AI) to digital breast tomosynthesis (DBT) are the reduction of reading times, the increase of diagnostic performance, and the reduction of interval cancer rates. In this review, after outlining the journey from computer-aided detection/diagnosis systems to AI applied to digital mammography (DM), we summarize the results of studies where AI was applied to DBT, noting that long-term advantages of DBT screening and its crucial ability to decrease the interval cancer rate are still under scrutiny. AI has shown the capability to overcome some shortcomings of DBT in the screening setting by improving diagnostic performance and by reducing recall rates (from -2 % to -27 %) and reading times (up to -53 %, with an average 20 % reduction), but the ability of AI to reduce interval cancer rates has not yet been clearly investigated. Prospective validation is needed to assess the cost-effectiveness and real-world impact of AI models assisting DBT interpretation, especially in large-scale studies with low breast cancer prevalence. Finally, we focus on the incoming era of personalized and risk-stratified screening that will first see the application of contrast-enhanced breast imaging to screen women with extremely dense breasts. As the diagnostic advantage of DBT over DM was concentrated in this category, we try to understand if the application of AI to DM in the remaining cohorts of women with heterogeneously dense or non-dense breast could close the gap in diagnostic performance between DM and DBT, thus neutralizing the usefulness of AI application to DBT.

Past, Present, and Future of Machine Learning and Artificial Intelligence for Breast Cancer Screening

Breast cancer screening has evolved substantially over the past few decades because of advancements in new image acquisition systems and novel artificial intelligence (AI) algorithms. This review provides a brief overview of the history, current state, and future of AI in breast cancer screening and diagnosis along with challenges involved in the development of AI systems. Although AI has been developing for interpretation tasks associated with breast cancer screening for decades, its potential to combat the subjective nature and improve the efficiency of human image interpretation is always expanding. The rapid advancement of computational power and deep learning has increased greatly in AI research, with promising performance in detection and classification tasks across imaging modalities. Most AI systems, based on human-engineered or deep learning methods, serve as concurrent or secondary readers, that is, as aids to radiologists for a specific, well-defined task. In the future, AI may be able to perform multiple integrated tasks, making decisions at the level of or surpassing the ability of humans. Artificial intelligence may also serve as a partial primary reader to streamline ancillary tasks, triaging cases or ruling out obvious normal cases. However, before AI is used as an independent, autonomous reader, various challenges need to be addressed, including explainability and interpretability, in addition to repeatability and generalizability, to ensure that AI will provide a significant clinical benefit to breast cancer screening across all populations.

Digital breast tomosynthesis-based peritumoral radiomics approaches in the differentiation of benign and malignant breast lesions

PURPOSE We aimed to evaluate digital breast tomosynthesis (DBT)-based radiomics in the differentiation of benign and malignant breast lesions in women. METHODS A total of 185 patients who underwent DBT scans were enrolled between December 2017 and June 2019. The features of handcrafted and deep learning-based radiomics were extracted from the tumoral and peritumoral regions with different radial dilation distances outside the tumor. A 3-step method was used to select discriminative features and develop the radiomics signature. Discriminative clinical factors were identified by univariate logistic regression. The clinical fac- tors with P < .05 were used to build a clinical model with multivariate logistic regression. The radiomics nomogram was developed by integrating the radiomics signature and discriminative clinical factors. Discriminative performance of the radiomics signature, clinical model, nomo- gram, and breast imaging reporting and data system assessment were evaluated and compared with the receiver operating characteristic and decision curves analysis (DCA). RESULTS A total of 2 handcrafted and 2 deep features were identified as the most discriminative features from the peritumoral regions with 2 mm dilation distances and used to develop the radiomics signature. The nomogram incorporating the radiomics signature, age, and menstruation status showed the best discriminative performance with area under the curve (AUC) values of 0.980 (95% CI, 0.960 to 1.000; sensitivity =0.970, specificity =0.946) in the training cohort and 0.985 (95% CI, 0.960 to 1.000; sensitivity = 0.909, specificity = 0.966) in the validation cohort. DCA con- firmed the potential clinical usefulness of our nomogram. CONCLUSION Our results illustrate that the radiomics nomogram integrating the DBT imaging features and clinical factors (age and menstruation status) can be considered as a useful tool in aiding the clinical diagnosis of breast cancer.

Automated Segmentation of Mass Regions in DBT Images Using a Dilated DCNN Approach

To overcome the limitations of conventional breast screening methods based on digital mammography, a quasi-3D imaging technique, digital breast tomosynthesis (DBT) has been developed in the field of breast cancer screening in recent years. In this work, a computer-aided architecture for mass regions segmentation in DBT images using a dilated deep convolutional neural network (DCNN) is developed. First, to improve the low contrast of breast tumour candidate regions and depress the background tissue noise in the DBT image effectively, the constraint matrix is established after top-hat transformation and multiplied with the DBT image. Second, input image patches are generated, and the data augmentation technique is performed to create the training data set for training a dilated DCNN architecture. Then the mass regions in DBT images are preliminarily segmented; each pixel is divided into two different kinds of labels. Finally, the postprocessing procedure removes all false-positives regions with less than 50 voxels. The final segmentation results are obtained by smoothing the boundaries of the mass regions with a median filter. The testing accuracy (ACC), sensitivity (SEN), and the area under the receiver operating curve (AUC) are adopted as the evaluation metrics, and the ACC, SEN, as well as AUC are 86.3%, 85.6%, and 0.852 for segmenting the mass regions in DBT images on the entire data set, respectively. The experimental results indicate that our proposed approach achieves promising results compared with other classical CAD-based frameworks.

A computer-aided approach for automatic detection of breast masses in digital mammogram via spectral clustering and support vector machine

Breast cancer continues to be a widespread health concern all over the world. Mammography is an important method in the early detection of breast abnormalities. In recent years, using an automatic Computer-Aided Detection (CAD) system based on image processing techniques has been a more reliable interpretation in the illustration of breast distortion. In this study, a fully process-integrated approach with developing a CAD system is presented for the detection of breast masses based on texture description, spectral clustering, and Support Vector Machine (SVM). To this end, breast Regions of Interest (ROIs) are automatically detected from digital mammograms via gray-scale enhancement and data cleansing. The ROIs are segmented as labeled multi-sectional patterns using spectral clustering by the means of intensity descriptors relying on the region's histogram and texture descriptors based on the Gray Level Co-occurrence Matrix (GLCM). In the next step, shape and probabilistic features are derived from the segmented sections and given to the Genetic Algorithm (GA) to do the feature selection. The optimal feature vector comprising a fusion of selected shape and probabilistic features is submitted to linear kernel SVM for robust and reliable classification of mass tissues from the non-mass. Linear discrimination analysis (LDA) is also performed to ascertain the significance of the nominated feature space. The classification results of the proposed approach are presented by sensitivity, specificity, and accuracy measures, which are 89.5%, 91.2%, and 90%, respectively.

Detecting Asymmetric Patterns and Localizing Cancers on Mammograms

One in eight women develops invasive breast cancer in her lifetime. The frontline protection against this disease is mammography. While computer-assisted diagnosis algorithms have made great progress in generating reliable global predictions, few focus on simultaneously producing regions of interest (ROIs) for biopsy. Can we combine ROI-oriented algorithms with global classification of cancer status, which simultaneously highlight suspicious regions and optimize classification performance? Can the asymmetry of breasts be adopted in deep learning for finding lesions and classifying cancers? We answer the above questions by building deep-learning networks that identify masses and microcalcifications in paired mammograms, exclude false positives, and stepwisely improve performance of the model with asymmetric information regarding the breasts. This method achieved a co-leading place in the Digital Mammography DREAM Challenge for predicting breast cancer. We highlight here the importance of this dual-purpose process that simultaneously provides the locations of potential lesions in mammograms.

•

A top-performing algorithm in the International DREAM Digital Mammography Challenge

•

Shows efficacy of asymmetric information from opposite breasts in identifying cancers

•

Integrated pixel-level localization and overall classification into the same software

Breast cancer affects one out of eight women in their lifetime. Given the importance of the need, in this work we present a region-of-interest-oriented deep-learning pipeline for detecting and locating breast cancers based on digital mammograms. It is a leading algorithm in the well-received Digital Mammography DREAM Challenge, in which computational methods were evaluated on large-scale, held-out testing sets of digital mammograms. This algorithm connects two aims: (1) determining whether a breast has cancer and (2) determining cancer-associated regions of interest. Particularly, we addressed the challenge of variation of mammogram images across different patients by pairing up the two opposite breasts to examine asymmetry, which substantially improved global classification as well as local lesion detection. We have dockerized this code, envisioning that it will be widely used in practice and as a future reference for digital mammography analysis.

Entropy-Based Breast Cancer Detection in Digital Mammograms Using World Cup Optimization Algorithm

Breast cancer is one of the deadliest cancers for women. Early detection of skin cancer gives a high chance for the women to escape from the malady and obtain a cure at the initial stages. In other words, early detection of breast cancer has a direct relation by the women's quality of life. In this case, mammography images are important. Indeed, the main test used for screening and early diagnosis of breast cancer is mammography. In recent years, computer-aided cancer detection has been turned into an active field of research and showed a promising future. In this study, a new optimization algorithm based on thresholding is introduced. A WCO algorithm is employed as the optimization algorithm. WCO is a new meta-heuristic approach which is inspired by the FIFA world cup challenge. The presented method utilizes random samples as candidate solutions from the search space inside the image histogram with considering to the objective function that is utilized by the Kapur's method.

DBT Masses Automatic Segmentation Using U-Net Neural Networks

To improve the automatic segmentation accuracy of breast masses in digital breast tomosynthesis (DBT) images, we propose a DBT mass automatic segmentation algorithm by using a U-Net architecture. Firstly, to suppress the background tissue noise and enhance the contrast of the mass candidate regions, after the top-hat transform of DBT images, a constraint matrix is constructed and multiplied with the DBT image. Secondly, an efficient U-Net neural network is built and image patches are extracted before data augmentation to establish the training dataset to train the U-Net model. And then the presegmentation of the DBT tumors is implemented, which initially classifies per pixel into two different types of labels. Finally, all regions smaller than 50 voxels considered as false positives are removed, and the median filter smoothes the mass boundaries to obtain the final segmentation results. The proposed method can effectively improve the performance in the automatic segmentation of the masses in DBT images. Using the detection Accuracy (Acc), Sensitivity (Sen), Specificity (Spe), and area under the curve (AUC) as evaluation indexes, the Acc, Sen, Spe, and AUC for DBT mass segmentation in the entire experimental dataset is 0.871, 0.869, 0.882, and 0.859, respectively. Our proposed U-Net-based DBT mass automatic segmentation system obtains promising results, which is superior to some classical architectures, and may be expected to have clinical application prospects.

A collection input based support tensor machine for lesion malignancy classification in digital breast tomosynthesis

Digital breast tomosynthesis (DBT) with improved lesion conspicuity and characterization has been adopted in screening practice. DBT-based diagnosis strongly depends on physicians' experience, so an automatic lesion malignancy classification model using DBT could improve the consistency of diagnosis among different physicians. Tensor-based approaches that use the original imaging data as input have shown promising results for many classification tasks. However, DBT data are pseudo-3D volumetric imaging as the slice spacing of DBT is much coarser than that of the in-plane resolution. Thus, directly constructing DBT as the third-order tensor in a conventional tensor-based classifier with introducing additional information to the original DBT data along the slice-spacing dimension will lead to inconsistency across all three dimensions. To avoid such inconsistency, we introduce a collection input based support tensor machine (CISTM)-based classifier that uses the tensor collection as input for classifying lesion malignancy in DBT. In CISTM, instead of introducing the third dimension directly into the geometry construction, the third-dimension structural relationship is related by weight parameters in the decision function, which is dynamically and automatically constructed during the classifier training process and is more consistent with the pseudo-3D nature of DBT. We tested our method on a DBT dataset of 926 images among which 262 were malignant and 664 were benign. We compared our method with the latest tensor-based method, KSTM (kernelled support tensor machine), which does not consider the unique non-uniform resolution property of DBT. Experimental results illustrate that the CISTM-based classifier is effective for classifying breast lesion malignancy in DBT and that it outperforms the KSTM-based classifier.

Transfer Learning From Convolutional Neural Networks for Computer-Aided Diagnosis: A Comparison of Digital Breast Tomosynthesis and Full-Field Digital Mammography

RATIONALE AND OBJECTIVES

With the growing adoption of digital breast tomosynthesis (DBT) in breast cancer screening, we compare the performance of deep learning computer-aided diagnosis on DBT images to that of conventional full-field digital mammography (FFDM).

MATERIALS AND METHODS

In this study, we retrospectively collected FFDM and DBT images of 78 biopsy-proven lesions from 76 patients. A region of interest was selected for each lesion on FFDM, synthesized 2D, and DBT key slice images. Features were extracted from each lesion using a pretrained convolutional neural network (CNN) and served as input to a support vector machine classifier trained in the task of predicting likelihood of malignancy.

RESULTS

From receiver operating characteristic (ROC) analysis of all 78 lesions, the synthesized 2D image performed best in both the cradiocaudal view (area under the ROC curve [AUC] = 0.81, SE = 0.05) and mediolateral oblique view (AUC = 0.88, SE = 0.04) in the task of lesion characterization. When cradiocaudal and mediolateral oblique data of each lesion were merged through soft voting, DBT key slice image performed best (AUC = 0.89, SE = 0.04). When only masses and architectural distortions (ARDs) were considered, DBT performed significantly better than FFDM (p = 0.024).

CONCLUSION

DBT performed significantly better than FFDM in the merged view classification of mass and ARD lesions. The increased performance suggests that the information extracted by the CNN from DBT images may be more relevant to lesion malignancy status than the information extracted from FFDM images. Therefore, this study provides supporting evidence for the efficacy of computer-aided diagnosis on DBT in the evaluation of mass and ARD lesions.

Breast Cancer Diagnosis in Digital Breast Tomosynthesis: Effects of Training Sample Size on Multi-Stage Transfer Learning Using Deep Neural Nets

In this paper, we developed a deep convolutional neural network (CNN) for the classification of malignant and benign masses in digital breast tomosynthesis (DBT) using a multi-stage transfer learning approach that utilized data from similar auxiliary domains for intermediate-stage fine-tuning. Breast imaging data from DBT, digitized screen-film mammography, and digital mammography totaling 4039 unique regions of interest (1797 malignant and 2242 benign) were collected. Using cross validation, we selected the best transfer network from six transfer networks by varying the level up to which the convolutional layers were frozen. In a single-stage transfer learning approach, knowledge from CNN trained on the ImageNet data was fine-tuned directly with the DBT data. In a multi-stage transfer learning approach, knowledge learned from ImageNet was first fine-tuned with the mammography data and then fine-tuned with the DBT data. Two transfer networks were compared for the second-stage transfer learning by freezing most of the CNN structures versus freezing only the first convolutional layer. We studied the dependence of the classification performance on training sample size for various transfer learning and fine-tuning schemes by varying the training data from 1% to 100% of the available sets. The area under the receiver operating characteristic curve (AUC) was used as a performance measure. The view-based AUC on the test set for single-stage transfer learning was 0.85 ± 0.05 and improved significantly (p <; 0.05$ ) to 0.91 ± 0.03 for multi-stage learning. This paper demonstrated that, when the training sample size from the target domain is limited, an additional stage of transfer learning using data from a similar auxiliary domain is advantageous.

Digital Breast Tomosynthesis: Physics, Artifacts, and Quality Control Considerations

As digital breast tomosynthesis (DBT) becomes widely used, radiologists must understand the basic principles of (a) image acquisition, (b) artifacts, and (c) quality control (QC) that are specific to DBT. Standard acquisition parameters common to both full-field digital mammography (FFDM) and DBT are combinations of x-ray tube voltage, current exposure time, and anode target and filter combinations. Image acquisition parameters specific to DBT include tube motion, sweep angle, and number of projections. Continuous tube motion or x-ray emission decreases imaging time but leads to focal spot blurring when compared with step-and-shoot techniques. The sweep angle and number of projections determines resolution. Wider sweep angles allow greater out-of-plane (z-axis) resolution, improving visualization of masses and architecture distortion. A greater number of projections increases in-plane or x-y axis resolution, improving visualization of microcalcifications. Artifacts related to DBT include blurring-ripple, truncation, and loss of skin and superficial tissue resolution. Motion artifacts are difficult to recognize because of inherent out-of-plane blurring. To maintain optimal image quality and an "as low as reasonably achievable" (ALARA) radiation dose, regular QC must be performed. DBT is considered a new imaging modality; therefore, breast imaging facilities are required to obtain a separate certification in addition to that in FFDM, and all personnel (radiologists, technologists, and medical physicists) are mandated to complete initial DBT training and maintain appropriate continuing medical education credits. ^©RSNA, 2019.

Computer-aided detection of mass in digital breast tomosynthesis using a faster region-based convolutional neural network

Digital breast tomosynthesis (DBT) is a newly developed three-dimensional tomographic imaging modality in the field of breast cancer screening designed to alleviate the limitations of conventional digital mammography-based breast screening methods. A computer-aided detection (CAD) system was designed for masses in DBT using a faster region-based convolutional neural network (faster-RCNN). To this end, a data set was collected, including 89 patients with 105 masses. An efficient detection architecture of convolution neural network with a region proposal network (RPN) was used for each slice to generate region proposals (i.e., bounding boxes) with a mass likelihood score. In each DBT volume, a slice fusion procedure was used to merge the detection results on consecutive 2D slices into one 3D DBT volume. The performance of the CAD system was evaluated using free-response receiver operating characteristic (FROC) curves. Our RCNN-based CAD system was compared with a deep convolutional neural network (DCNN)-based CAD system. The RCNN-based CAD generated a performance with an area under the ROC (AUC) of 0.96, whereas the DCNN-based CAD achieved a performance with AUC of 0.92. For lesion-based mass detection, the sensitivity of RCNN-based CAD was 90% at 1.54 false positive (FP) per volume, whereas the sensitivity of DCNN-based CAD was 90% at 2.81 FPs/volume. For breast-based mass detection, RCNN-based CAD generated a sensitivity of 90% at 0.76 FP/breast, which is significantly increased compared with the DCNN-based CAD with a sensitivity of 90% at 2.25 FPs/breast. The results suggest that the faster R-CNN has the potential to augment the prescreening and FP reduction in the CAD system for masses.

Comparison of Digital Breast Tomosynthesis and Digital Mammography for Detection of Breast Cancer in Kuwaiti Women

OBJECTIVE

To investigate the sensitivity and specificity of digital mammography (DM) and digital breast tomosynthesis (DBT) for the detection of breast cancer in comparison to histopathology findings.

SUBJECTS AND METHODS

We included 65 breast lesions in 58 women, each detected by two diagnostic mammography techniques - DM and DBT using Senographe Essential (GE Healthcare, Buc, France) - and subsequently confirmed by histopathology. The Breast Imaging Reporting and Data System was used for characterizing the lesions.

RESULTS

The average age of women was 48.3 years (range 26-81 years). There were 34 malignant and 31 benign breast lesions. The sensitivity of DM and DBT was 73.5 and 100%, respectively, while the specificity was 67.7 and 94%, respectively. Receiver operating characteristic curve analysis showed an overall diagnostic advantage of DBT over DM, with a significant difference between DBT and DM (p < 0.001). By performing Cohen's kappa test, we found that there was a strong level of agreement according to Altman guidelines between DBT and histopathology findings (0.97), but a weak agreement between DM and histopathology findings (0.47).

CONCLUSION

DBT improves the clinical accuracy of mammography by increasing both sensitivity and specificity. We believe that this improvement is due to improved image visibility and quality. These results could be of interest to health care institutions as they may impact their decision on whether to upgrade to DBT not only for diagnosis, but also for screening.

Application of a pixel-shifted linear interpolation technique for reducing the projection number in tomosynthesis imaging

Tomosynthesis images are reconstructed from several projections. However, the number of projections is proportional to the exposure dose, and a reduction in the number of projections would result in a reduction of radiation dose to the patient but also degradation of image quality. The purpose of this study was to propose a new computerized method to supply interpolation images instead of real projection images for maintaining the number of projection images and image quality of reconstructed tomosynthesis images. A set of images comprising one-half the number of projection images [37 projections (Half set)], selected from the original full set of projection images [73 projections (Full set)], was used at an interval of one by one. In this study, the authors used a new linear interpolation technique (Shift-Linear method), which takes into account shifted distances between two corresponding pixels on two projection images. The image quality of tomosynthesis images reconstructed from the full set and the quasi-full projection images, which were produced from the Half set using the Shift-Linear method, was compared. Image quality was assessed in terms of modulation transfer function, noise power spectrum, contrast-to-noise ratio, and the detective quantum efficiency. Using this proposed method, the image quality of reconstructed tomosynthesis images could be maintained with the reduction of approximately 50% exposure dose.

A review of computer aided detection in mammography

Breast screening with mammography is widely recognized as the most effective method of detecting early breast cancer and has consistently demonstrated a 20-40% decrease in mortality among screened women. Despite this, the sensitivity of mammography ranges between 70 and 90%. Computer aided detection (CAD) is an artificial intelligence (AI) technique that utilizes pattern recognition to highlight suspicious features on imaging and marks them for the radiologist to review and interpret. It aims to decrease oversights made by interpreting radiologists. Here we review the efficacy of CAD and potential future directions.

Mass detection in digital breast tomosynthesis data using convolutional neural networks and multiple instance learning

Digital breast tomosynthesis (DBT) was developed in the field of breast cancer screening as a new tomographic technique to minimize the limitations of conventional digital mammography breast screening methods. A computer-aided detection (CAD) framework for mass detection in DBT has been developed and is described in this paper. The proposed framework operates on a set of two-dimensional (2D) slices. With plane-to-plane analysis on corresponding 2D slices from each DBT, it automatically learns complex patterns of 2D slices through a deep convolutional neural network (DCNN). It then applies multiple instance learning (MIL) with a randomized trees approach to classify DBT images based on extracted information from 2D slices. This CAD framework was developed and evaluated using 5040 2D image slices derived from 87 DBT volumes. The empirical results demonstrate that this proposed CAD framework achieves much better performance than CAD systems that use hand-crafted features and deep cardinality-restricted Bolzmann machines to detect masses in DBTs.

Multimodal breast cancer imaging using coregistered dynamic diffuse optical tomography and digital breast tomosynthesis

Diffuse optical tomography (DOT) is emerging as a noninvasive functional imaging method for breast cancer diagnosis and neoadjuvant chemotherapy monitoring. In particular, the multimodal approach of combining DOT with x-ray digital breast tomosynthesis (DBT) is especially synergistic as DBT prior information can be used to enhance the DOT reconstruction. DOT, in turn, provides a functional information overlay onto the mammographic images, increasing sensitivity and specificity to cancer pathology. We describe a dynamic DOT apparatus designed for tight integration with commercial DBT scanners and providing a fast (up to 1 Hz) image acquisition rate to enable tracking hemodynamic changes induced by the mammographic breast compression. The system integrates 96 continuous-wave and 24 frequency-domain source locations as well as 32 continuous wave and 20 frequency-domain detection locations into low-profile plastic plates that can easily mate to the DBT compression paddle and x-ray detector cover, respectively. We demonstrate system performance using static and dynamic tissue-like phantoms as well as in vivo images acquired from the pool of patients recalled for breast biopsies at the Massachusetts General Hospital Breast Imaging Division.

Mass detection in digital breast tomosynthesis: Deep convolutional neural network with transfer learning from mammography

PURPOSE

Develop a computer-aided detection (CAD) system for masses in digital breast tomosynthesis (DBT) volume using a deep convolutional neural network (DCNN) with transfer learning from mammograms.

METHODS

A data set containing 2282 digitized film and digital mammograms and 324 DBT volumes were collected with IRB approval. The mass of interest on the images was marked by an experienced breast radiologist as reference standard. The data set was partitioned into a training set (2282 mammograms with 2461 masses and 230 DBT views with 228 masses) and an independent test set (94 DBT views with 89 masses). For DCNN training, the region of interest (ROI) containing the mass (true positive) was extracted from each image. False positive (FP) ROIs were identified at prescreening by their previously developed CAD systems. After data augmentation, a total of 45 072 mammographic ROIs and 37 450 DBT ROIs were obtained. Data normalization and reduction of non-uniformity in the ROIs across heterogeneous data was achieved using a background correction method applied to each ROI. A DCNN with four convolutional layers and three fully connected (FC) layers was first trained on the mammography data. Jittering and dropout techniques were used to reduce overfitting. After training with the mammographic ROIs, all weights in the first three convolutional layers were frozen, and only the last convolution layer and the FC layers were randomly initialized again and trained using the DBT training ROIs. The authors compared the performances of two CAD systems for mass detection in DBT: one used the DCNN-based approach and the other used their previously developed feature-based approach for FP reduction. The prescreening stage was identical in both systems, passing the same set of mass candidates to the FP reduction stage. For the feature-based CAD system, 3D clustering and active contour method was used for segmentation; morphological, gray level, and texture features were extracted and merged with a linear discriminant classifier to score the detected masses. For the DCNN-based CAD system, ROIs from five consecutive slices centered at each candidate were passed through the trained DCNN and a mass likelihood score was generated. The performances of the CAD systems were evaluated using free-response ROC curves and the performance difference was analyzed using a non-parametric method.

RESULTS

Before transfer learning, the DCNN trained only on mammograms with an AUC of 0.99 classified DBT masses with an AUC of 0.81 in the DBT training set. After transfer learning with DBT, the AUC improved to 0.90. For breast-based CAD detection in the test set, the sensitivity for the feature-based and the DCNN-based CAD systems was 83% and 91%, respectively, at 1 FP/DBT volume. The difference between the performances for the two systems was statistically significant (p-value < 0.05).

CONCLUSIONS

The image patterns learned from the mammograms were transferred to the mass detection on DBT slices through the DCNN. This study demonstrated that large data sets collected from mammography are useful for developing new CAD systems for DBT, alleviating the problem and effort of collecting entirely new large data sets for the new modality.

Automated Detection Algorithm of Breast Masses in Three-Dimensional Ultrasound Images

Objectives

We propose an automatic breast mass detection algorithm in three-dimensional (3D) ultrasound (US) images using the Hough transform technique.

Methods

One hundred twenty-five cropped images containing 68 benign and 60 malignant masses are acquired with clinical diagnosis by an experienced radiologist. The 3D US images are masked, subsampled, contrast-adjusted, and median-filtered as preprocessing steps before the Hough transform is used. Thereafter, we perform 3D Hough transform to detect spherical hyperplanes in 3D US breast image volumes, generate Hough spheres, and sort them in the order of votes. In order to reduce the number of the false positives in the breast mass detection algorithm, the Hough sphere with a mean or grey level value of the centroid higher than the mean of the 3D US image is excluded, and the remaining Hough sphere is converted into a circumscribing parallelepiped cube as breast mass lesion candidates. Finally, we examine whether or not the generated Hough cubes were overlapping each other geometrically, and the resulting Hough cubes are suggested as detected breast mass candidates.

Results

An automatic breast mass detection algorithm is applied with mass detection sensitivity of 96.1% at 0.84 false positives per case, quite comparable to the results in previous research, and we note that in the case of malignant breast mass detection, every malignant mass is detected with false positives per case at a rate of 0.62.

Conclusions

The breast mass detection efficiency of our algorithm is assessed by performing a ROC analysis.

Three-Dimensional Computer-Aided Detection of Microcalcification Clusters in Digital Breast Tomosynthesis

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

Digital Breast Tomosynthesis: State of the Art

This topical review on digital breast tomosynthesis (DBT) is provided with the intent of describing the state of the art in terms of technology, results from recent clinical studies, advanced applications, and ongoing efforts to develop multimodality imaging systems that include DBT. Particular emphasis is placed on clinical studies. The observations of increase in cancer detection rates, particularly for invasive cancers, and the reduction in false-positive rates with DBT in prospective trials indicate its benefit for breast cancer screening. Retrospective multireader multicase studies show either noninferiority or superiority of DBT compared with mammography. Methods to curtail radiation dose are of importance. (©) RSNA, 2015.

Digital breast tomosynthesis: lessons learned from early clinical implementation

The limitations of mammography are well known and are partly related to the fact that with conventional imaging, the three-dimensional volume of the breast is imaged and presented in a two-dimensional format. Because normal breast tissue is similar in x-ray attenuation to some breast cancers, clinically relevant malignancies may be obscured by normal overlapping tissue. In addition, complex areas of normal tissue may be perceived as suspicious. The limitations of two-dimensional breast imaging lead to low sensitivity in detecting some cancers and high false-positive recall rates. Although mammographic screening has been shown to reduce breast cancer deaths by approximately 30%, controversy exists over when and how often screening mammography should occur. Digital breast tomosynthesis (DBT) is rapidly being implemented in breast imaging clinics around the world as early clinical data demonstrate that it may address some of the limitations of conventional mammography. With DBT, multiple low-dose x-ray images are acquired in an arc and reconstructed to create a three-dimensional image, thus minimizing the impact of overlapping breast tissue and improving lesion conspicuity. Early studies of screening DBT have shown decreased false-positive callback rates and increased rates of cancer detection (particularly for invasive cancers), resulting in increased sensitivity and specificity. In our clinical practice, we have completed more than 2 years of using two-view digital mammography combined with two-view DBT for all screening and select diagnostic imaging examinations (over 25,000 patients). Our experience, combined with previously published data, demonstrates that the combined use of DBT and digital mammography is associated with improved outcomes for screening and diagnostic imaging. Online supplemental material is available for this article.

Breast MRI, digital mammography and breast tomosynthesis: comparison of three methods for early detection of breast cancer

Breast cancer is the most common malignancy in women and early detection is important for its successful treatment. The aim of this study was to investigate the sensitivity and specificity of three methods for early detection of breast cancer: breast magnetic resonance imaging (MRI), digital mammography, and breast tomosynthesis in comparison to histopathology, as well as to investigate the intraindividual variability between these modalities. We included 57 breast lesions, each detected by three diagnostic modalities: digital mammography, breast MRI, and breast tomosynthesis, and subsequently confirmed by histopathology. Breast Imaging-Reporting and Data System (BI-RADS) was used for characterizing the lesions. One experienced radiologist interpreted all three diagnostic modalities. Twenty-nine of the breast lesions were malignant while 28 were benign. The sensitivity for digital mammography, breast MRI, and breast tomosynthesis, was 72.4%, 93.1%, and 100%, respectively; while the specificity was 46.4%, 60.7%, and 75%, respectively. Receiver operating characteristics (ROC) curve analysis showed an overall diagnostic advantage of breast tomosynthesis over both breast MRI and digital mammography. The difference in performance between breast tomosynthesis and digital mammography was significant (p <0.001), while the difference between breast tomosynthesis and breast MRI was not significant (p=0.20).

Wavelet-based 3D reconstruction of microcalcification clusters from two mammographic views: new evidence that fractal tumors are malignant and Euclidean tumors are benign

The 2D Wavelet-Transform Modulus Maxima (WTMM) method was used to detect microcalcifications (MC) in human breast tissue seen in mammograms and to characterize the fractal geometry of benign and malignant MC clusters. This was done in the context of a preliminary analysis of a small dataset, via a novel way to partition the wavelet-transform space-scale skeleton. For the first time, the estimated 3D fractal structure of a breast lesion was inferred by pairing the information from two separate 2D projected mammographic views of the same breast, i.e. the cranial-caudal (CC) and mediolateral-oblique (MLO) views. As a novelty, we define the “CC-MLO fractal dimension plot”, where a “fractal zone” and “Euclidean zones” (non-fractal) are defined. 118 images (59 cases, 25 malignant and 34 benign) obtained from a digital databank of mammograms with known radiologist diagnostics were analyzed to determine which cases would be plotted in the fractal zone and which cases would fall in the Euclidean zones. 92% of malignant breast lesions studied (23 out of 25 cases) were in the fractal zone while 88% of the benign lesions were in the Euclidean zones (30 out of 34 cases). Furthermore, a Bayesian statistical analysis shows that, with 95% credibility, the probability that fractal breast lesions are malignant is between 74% and 98%. Alternatively, with 95% credibility, the probability that Euclidean breast lesions are benign is between 76% and 96%. These results support the notion that the fractal structure of malignant tumors is more likely to be associated with an invasive behavior into the surrounding tissue compared to the less invasive, Euclidean structure of benign tumors. Finally, based on indirect 3D reconstructions from the 2D views, we conjecture that all breast tumors considered in this study, benign and malignant, fractal or Euclidean, restrict their growth to 2-dimensional manifolds within the breast tissue.

Update on new technologies in digital mammography

Despite controversy regarding mammography's efficacy, it continues to be the most commonly used breast cancer-screening modality. With the development of digital mammography, some improved benefit has been shown in women with dense breast tissue. However, the density of breast tissue continues to limit the sensitivity of conventional mammography. We discuss the development of some derivative digital technologies, primarily digital breast tomosynthesis, and their strengths, weaknesses, and potential patient impact.

Breast mass detection using slice conspicuity in 3D reconstructed digital breast volumes

In digital breast tomosynthesis, the three dimensional (3D) reconstructed volumes only provide quasi-3D structure information with limited resolution along the depth direction due to insufficient sampling in depth direction and the limited angular range. The limitation could seriously hamper the conventional 3D image analysis techniques for detecting masses because the limited number of projection views causes blurring in the out-of-focus planes. In this paper, we propose a novel mass detection approach using slice conspicuity in the 3D reconstructed digital breast volumes to overcome the above limitation. First, to overcome the limited resolution along the depth direction, we detect regions of interest (ROIs) on each reconstructed slice and separately utilize the depth directional information to combine the ROIs effectively. Furthermore, we measure the blurriness of each slice for resolving the degradation of performance caused by the blur in the out-of-focus plane. Finally, mass features are extracted from the selected in focus slices and analyzed by a support vector machine classifier to reduce the false positives. Comparative experiments have been conducted on a clinical data set. Experimental results demonstrate that the proposed approach outperforms the conventional 3D approach by achieving a high sensitivity with a small number of false positives.

Integration of microwave tomography with magnetic resonance for improved breast imaging

PURPOSE

Breast magnetic resonance imaging is highly sensitive but not very specific for the detection of breast cancer. Opportunities exist to supplement the image acquisition with a more specific modality provided the technical challenges of meeting space limitations inside the bore, restricted breast access, and electromagnetic compatibility requirements can be overcome. Magnetic resonance (MR) and microwave tomography (MT) are complementary and synergistic because the high resolution of MR is used to encode spatial priors on breast geometry and internal parenchymal features that have distinct electrical properties (i.e., fat vs fibroglandular tissue) for microwave tomography.

METHODS

The authors have overcome integration challenges associated with combining MT with MR to produce a new coregistered, multimodality breast imaging platform--magnetic resonance microwave tomography, including: substantial illumination tank size reduction specific to the confined MR bore diameter, minimization of metal content and composition, reduction of metal artifacts in the MR images, and suppression of unwanted MT multipath signals.

RESULTS

MR SNR exceeding 40 dB can be obtained. Proper filtering of MR signals reduces MT data degradation allowing MT SNR of 20 dB to be obtained, which is sufficient for image reconstruction. When MR spatial priors are incorporated into the recovery of MT property estimates, the errors between the recovered versus actual dielectric properties approach 5%.

CONCLUSIONS

The phantom and human subject exams presented here are the first demonstration of combining MT with MR to improve the accuracy of the reconstructed MT images.

Clinical implementation of digital breast tomosynthesis

Digital breast tomosynthesis is rapidly being implemented in breast imaging clinics across the world as early clinical data demonstrate that this innovative technology may address some of the long-standing limitations of conventional mammography. This article reviews the recent clinical data supporting digital breast tomosynthesis implementation, the basics of digital breast tomosynthesis image interpretation using case-based illustrations, and potential issues to consider as this new technology is integrated into daily clinical use.

Breast image analysis for risk assessment, detection, diagnosis, and treatment of cancer

The role of breast image analysis in radiologists' interpretation tasks in cancer risk assessment, detection, diagnosis, and treatment continues to expand. Breast image analysis methods include segmentation, feature extraction techniques, classifier design, biomechanical modeling, image registration, motion correction, and rigorous methods of evaluation. We present a review of the current status of these task-based image analysis methods, which are being developed for the various image acquisition modalities of mammography, tomosynthesis, computed tomography, ultrasound, and magnetic resonance imaging. Depending on the task, image-based biomarkers from such quantitative image analysis may include morphological, textural, and kinetic characteristics and may depend on accurate modeling and registration of the breast images. We conclude with a discussion of future directions.

Digital tomosynthesis: a new future for breast imaging?

The aim of this article is to review the major limitations in current mammography and to describe how these may be addressed by digital breast tomosynthesis (DBT). DBT is a novel imaging technology in which an x-ray fan beam sweeps in an arc across the breast, producing tomographic images and enabling the production of volumetric, three-dimensional (3D) data. It can reduce tissue overlap encountered in conventional two-dimensional (2D) mammography, and thus has the potential to improve detection of breast cancer, reduce the suspicious presentations of normal tissues, and facilitate accurate differentiation of lesion types. This paper reviews the latest studies of this new technology. Issues including diagnostic efficacy, reading time, radiation dose, and level of compression; cost and new innovations are considered.

A review of breast tomosynthesis. Part II. Image reconstruction, processing and analysis, and advanced applications

Many important post-acquisition aspects of breast tomosynthesis imaging can impact its clinical performance. Chief among them is the reconstruction algorithm that generates the representation of the three-dimensional breast volume from the acquired projections. But even after reconstruction, additional processes, such as artifact reduction algorithms, computer aided detection and diagnosis, among others, can also impact the performance of breast tomosynthesis in the clinical realm. In this two part paper, a review of breast tomosynthesis research is performed, with an emphasis on its medical physics aspects. In the companion paper, the first part of this review, the research performed relevant to the image acquisition process is examined. This second part will review the research on the post-acquisition aspects, including reconstruction, image processing, and analysis, as well as the advanced applications being investigated for breast tomosynthesis.

Computer-aided lesion diagnosis in automated 3-D breast ultrasound using coronal spiculation

A computer-aided diagnosis (CAD) system for the classification of lesions as malignant or benign in automated 3-D breast ultrasound (ABUS) images, is presented. Lesions are automatically segmented when a seed point is provided, using dynamic programming in combination with a spiral scanning technique. A novel aspect of ABUS imaging is the presence of spiculation patterns in coronal planes perpendicular to the transducer. Spiculation patterns are characteristic for malignant lesions. Therefore, we compute spiculation features and combine them with features related to echotexture, echogenicity, shape, posterior acoustic behavior and margins. Classification experiments were performed using a support vector machine classifier and evaluation was done with leave-one-patient-out cross-validation. Receiver operator characteristic (ROC) analysis was used to determine performance of the system on a dataset of 201 lesions. We found that spiculation was among the most discriminative features. Using all features, the area under the ROC curve (A(z)) was 0.93, which was significantly higher than the performance without spiculation features (A(z)=0.90, p=0.02). On a subset of 88 cases, classification performance of CAD (A(z)=0.90) was comparable to the average performance of 10 readers (A(z)=0.87).

Computer-aided detection of clustered microcalcifications in digital breast tomosynthesis: a 3D approach

PURPOSE

To design a computer-aided detection (CADe) system for clustered microcalcifications in reconstructed digital breast tomosynthesis (DBT) volumes and to perform a preliminary evaluation of the CADe system.

METHODS

IRB approval and informed consent were obtained in this study. A data set of two-view DBT of 72 breasts containing microcalcification clusters was collected from 72 subjects who were scheduled to undergo breast biopsy. Based on tissue sampling results, 17 cases had breast cancer and 55 were benign. A separate data set of two-view DBT of 38 breasts free of clustered microcalcifications from 38 subjects was collected to independently estimate the number of false-positives (FPs) generated by the CADe system. A radiologist experienced in breast imaging marked the biopsied cluster of microcalcifications with a 3D bounding box using all available clinical and imaging information. A CADe system was designed to detect microcalcification clusters in the reconstructed volume. The system consisted of prescreening, clustering, and false-positive reduction stages. In the prescreening stage, the conspicuity of microcalcification-like objects was increased by an enhancement-modulated 3D calcification response function. An iterative thresholding and 3D object growing method was used to detect cluster seed objects, which were used as potential centers of microcalcification clusters. In the cluster detection stage, microcalcification candidates were identified using a second iterative thresholding procedure, which was applied to the signal-to-noise ratio (SNR) enhanced image voxels with a positive calcification response. Starting with each cluster seed object as the initial cluster center, a dynamic clustering algorithm formed a cluster candidate by including microcalcification candidates within a 3D neighborhood of the cluster seed object that satisfied the clustering criteria. The number, size, and SNR of the microcalcifications in a cluster candidate and the cluster shape were used to reduce the number of FPs.

RESULTS

The prescreening stage detected a cluster seed object in 94% of the biopsied microcalcification clusters at a threshold of 100 cluster seed objects per DBT volume. After clustering, the detection sensitivity was 90% at 15 marks per DBT volume. After FP reduction, at 85% sensitivity, the average number of FPs estimated using the data set containing microcalcification clusters was 3.8 per DBT volume, and that estimated using the data set free of microcalcification clusters was 3.4. The detection performance for malignant microcalcification clusters was superior to that for benign clusters.

CONCLUSIONS

Our study indicates the feasibility of the 3D approach to the detection of clustered microcalcifications in DBT and that the newly designed enhancement-modulated 3D calcification response function is promising for prescreening. Further work is needed to assess the generalizability of our approach and to improve its performance.

Computer-Aided Detection and Diagnosis of Breast Cancer Using Machine Learning, Texture and Shape Features

Breast cancer is a malignant (cancer) tumor that starts from cells of the breast, being the major cause of deaths by cancer in the female population. There has been tremendous interest in the use of image processing and analysis techniques for computer aided detection (CAD)/ diagnostics (CADx) in digital mammograms. The goal has been to increase diagnostic accuracy as well as the reproducibility of mammographic interpretation. CAD/CADx systems can aid radiologists by providing a second opinion and may be used in the first stage of examination in the near future, providing the reduction of the variability among radiologists in the interpretation of mammograms. This chapter provides an overview of techniques used in computer-aided detection and diagnosis of breast cancer. The authors focus on the application of texture and shape tissues signature used with machine learning techniques, like support vector machines (SVM) and growing neural gas (GNG).

Diagnosis of lymph node metastasis in uterine cervical cancer: usefulness of computer-aided diagnosis with comprehensive evaluation of MR images and clinical findings

BACKGROUND

Lymph node (LN) status is an important parameter for determining the treatment strategy and for predicting the prognosis for patients with uterine cervical cancer. Computer-aided diagnosis (CAD) can be feasible for differentiating metastatic from non-metastatic lymph nodes in patients with uterine cervical cancer.

PURPOSE

To determine the usefulness of CAD that comprehensively evaluates MR images and clinical findings for detecting LN metastasis in uterine cervical cancer.

MATERIAL AND METHODS

In 680 LNs from 143 patients who underwent radical hysterectomy for uterine cervical cancer, the CAD system using the Bayesian classifier estimated the probability of metastasis based on MR findings and clinical findings. We compared the diagnostic accuracy for detecting metastatic LNs in the CAD and MR findings.

RESULTS

Metastasis was diagnosed in 70 (12%) LNs from 34 (24%) patients. The area under ROC curves of CAD (0.924) was greater than those of the mean ADC (0.854), minimum ADC (0.849), maximum ADC (0.827), short-axis diameter (0.856) and long-axis diameter (0.753) (P < 0.05). The specificity and accuracy of the CAD (86%, 86%) were greater than those of the mean ADC (77%, 77%), maximum ADC (77%, 77%), minimum ADC (68%, 70%), and short-axis diameter (65%, 67%) (P < 0.05).

CONCLUSION

CAD system can improve the diagnostic performance of MR for detecting metastatic LNs in uterine cervical cancer.

Breast tomosynthesis

Digital mammography has been well-evaluated for the diagnosis of breast cancer. The scientific data show that mammography alone, especially in dense breast parenchyma, has its weaknesses. These weaknesses are due to the low contrast of tumors in comparison with the surrounding parenchyma and the overlying structures that mask tumors. The initial results from tomosynthesis studies show a tendency for better imaging and higher accuracy and lower recall rates with tomosynthesis. We present in this article a literature review of the development of breast tomosynthesis and follow it with case examples.

Mutual information-based template matching scheme for detection of breast masses: from mammography to digital breast tomosynthesis

Development of a computational decision aid for a new medical imaging modality typically is a long and complicated process. It consists of collecting data in the form of images and annotations, development of image processing and pattern recognition algorithms for analysis of the new images and finally testing of the resulting system. Since new imaging modalities are developed more rapidly than ever before, any effort for decreasing the time and cost of this development process could result in maximizing the benefit of the new imaging modality to patients by making the computer aids quickly available to radiologists that interpret the images. In this paper, we make a step in this direction and investigate the possibility of translating the knowledge about the detection problem from one imaging modality to another. Specifically, we present a computer-aided detection (CAD) system for mammographic masses that uses a mutual information-based template matching scheme with intelligently selected templates. We presented principles of template matching with mutual information for mammography before. In this paper, we present an implementation of those principles in a complete computer-aided detection system. The proposed system, through an automatic optimization process, chooses the most useful templates (mammographic regions of interest) using a large database of previously collected and annotated mammograms. Through this process, the knowledge about the task of detecting masses in mammograms is incorporated in the system. Then, we evaluate whether our system developed for screen-film mammograms can be successfully applied not only to other mammograms but also to digital breast tomosynthesis (DBT) reconstructed slices without adding any DBT cases for training. Our rationale is that since mutual information is known to be a robust inter-modality image similarity measure, it has high potential of transferring knowledge between modalities in the context of the mass detection task. Experimental evaluation of the system on mammograms showed competitive performance compared to other mammography CAD systems recently published in the literature. When the system was applied "as-is" to DBT, its performance was notably worse than that for mammograms. However, with a simple additional preprocessing step, the performance of the system reached levels similar to that obtained for mammograms. In conclusion, the presented CAD system not only performed competitively on screen-film mammograms but it also performed robustly on DBT showing that direct transfer of knowledge across breast imaging modalities for mass detection is in fact possible.