Mammographic Computer-aided Diagnosis Research Articles

Computer‐aided diagnosis in the era of deep learning

Computer-aided diagnosis (CAD) has been a major field of research for the past few decades. CAD uses machine learning methods to analyze imaging and/or nonimaging patient data and makes assessment of the patient's condition, which can then be used to assist clinicians in their decision-making process. The recent success of the deep learning technology in machine learning spurs new research and development efforts to improve CAD performance and to develop CAD for many other complex clinical tasks. In this paper, we discuss the potential and challenges in developing CAD tools using deep learning technology or artificial intelligence (AI) in general, the pitfalls and lessons learned from CAD in screening mammography and considerations needed for future implementation of CAD or AI in clinical use. It is hoped that the past experiences and the deep learning technology will lead to successful advancement and lasting growth in this new era of CAD, thereby enabling CAD to deliver intelligent aids to improve health care.

Importance of Better Human-Computer Interaction in the Era of Deep Learning: Mammography Computer-Aided Diagnosis as a Use Case

Importance of Better Human-Computer Interaction in the Era of Deep Learning: Mammography Computer-Aided Diagnosis as a Use Case

WE‐DE‐207B‐02: Detection of Masses On Mammograms Using Deep Convolutional Neural Network: A Feasibility Study

Purpose:To develop a deep convolutional neural network (DCNN)‐based computer‐aided diagnosis (CAD) system for detecting the masses in digital mammographic images.Methods:A DCNN architecture, which consists of 5 convolutional layers and 3 fully connected layers, is constructed in this study. The DCNN parameters are then trained by the following two procedures. We first train the DCNN using about 1.3 million natural images for classification of 1,000 categories. Then, we modify the last fully connected layer and subsequently train the modified DCNN using 1,656 mammographic region of interest (ROI) images for two categories classification: mass and normal.Results:The trained DCNN is tested by using 198 mammographic ROI images including 99 mass images and 99 normal images. The experimental results show that the sensitivity of the mass detection is about 89.9% and the false positive is 19.2%. These results demonstrated that the DCNN has a potential for mammographic CAD.Conclusion:In recent years, the DCNN, as one of the most successful techniques in deep learning technology, made a remarkable impact on image recognition application. For medical image recognition, however, its performance is uncertainty because collecting a large amount of training image data for a particularly medical image modality is difficult. In this study, our preliminary experiments demonstrated a feasibility to apply the DCNN in mammographic CAD system. To the best of our knowledge, this study is also the first demonstration of DCNN for detecting the masses in mammographic images.

Latent feature mining of spatial and marginal characteristics for mammographic mass classification

Mass classification is one of the key procedures in mammography computer-aided diagnosis (CAD) system, which is widely applied to help improving clinic diagnosis performance. In literature, classical mass classification systems always employ a large number and types of features for discriminating masses. This will produce higher computational complexity. And the incompatibility among various features also may introduce some negative impact to classification accuracy. Furthermore, latent characteristics of masses are seldom considered in the present scheme, which are useful to reveal hidden distribution pattern of masses. For the above purpose, the paper proposes a new mammographic mass classification scheme. Mammograms are detected and segmented first for obtaining region of interests with masses (ROIms). Then Latent Dirichlet Allocation (LDA) is introduced to find hidden topic distribution of ROIms. A special spatial pyramid structure is proposed and incorporated with LDA for capturing latent spatial characteristics of ROIms. For mining latent statistical marginal characteristics of masses, local patches on segmented boundary are extracted to construct a special document for LDA. Finally, all the latent topics will be combined, analyzed and classified by employing the SVM classifier. The experimental results on a dataset in DDSM demonstrate the effectiveness and efficiency of the proposed classification scheme.

Fissures Segmentation Using Surface Features: Content-based Retrieval for Mammographic Mass Using Ensemble Classifier

Accurate classification is critical in mammography computer-aided diagnosis using content-based image retrieval approaches (CBIR CAD). The objectives of this study were to: 1) develop an accurate ensemble classifier based on domain knowledge and a robust feature selection method for CBIR CAD; 2) propose three new features; and 3) assess the performance of the proposed method and new features by using a relatively large imaging data set. The data set used in this study consisted of 2114 regions of interest (ROI) extracted from a publicly available image database. The proposed ensemble classifier method we called E-DGA-KNN included four steps. In the first step, 804 ROIs depict masses were divided into five classes according to their boundary types. Then, each class of ROI with an equal number of negative ROIs were put together to create a sub-database. Second, a dual-stage genetic algorithm, which was called DGA, was applied on those five sub-databases for feature selection and weights determination respectively. In the third step, five base K-nearest neighbor (KNN) classifiers were created by using the results of the second step on 2114 ROIs, and five detection scores for a given queried ROI were obtained. Finally, these classifiers are combined to yield a final classification. The performances of the proposed methods were evaluated by using receiver operating characteristic (ROC) analysis. A comparison with eight different methods on the data set was provided which include the stepwise linear discriminative analysis algorithm (SLDA) and particle swarm optimization (PSO) algorithm with KNN classifier. When four hybrid feature selection methods were applied with single KNN classifier (ie, DGA-KNN, SLDA-WGA-KNN, SLDA-PSO-KNN, GA-PSO-KNN) and the proposed E-DGA-KNN method to the data set, the computed areas under the ROC curve (Az) were 0.8782 ± 0.0080, 0.8675 ± 0.0081, 0.8623 ± 0.0083, 0.8725 ± 0.0079, and 0.8927 ± 0.0073, respectively. If all features and single KNN classifier were used, the Az value was 0.8478 ± 0.0088. Az values were 0.8592 ± 0.0083 and 0.8632 ± 0.0081 when SLDA or GA algorithm used alone. In this study, an ensemble classifier based on domain knowledge and a dual-stage feature selection method was proposed. Evaluation results indicated that the proposed method achieved largest value of ROC compared to other algorithms. The proposed method shows better performance and has the potential to improve the performance of CBIR CAD in interpreting and analyzing mammograms.

Breast masses detection using phase portrait analysis and fuzzy inference systems

Breast masses exhibit variability in margins, shapes, and dimensions, so their detection is a difficult task in mammographic computer-aided diagnosis. Mass detection is usually a two-step procedure: mass identification and false-positive reduction. A new method to automatically detect mass lesions in mammographic images with tuning according to the breast tissue density was developed and tested. A modified phase portrait analysis method was introduced, based on the eigenvalue condition number and an eigenvalue intensity map. The method uses an iterative and tissue density-adaptive segmentation procedure with extraction of geometric features. False-positive reduction is accomplished using a fuzzy inference-based classifier. A leave-one-image-out cross-validation procedure was implemented, and stepwise regression analysis was used to automatically extract an optimal set of features. Testing and validation were performed on two different data sets containing at least one malignant mass D1 (388 images) and D2 (674 images), and a third data set N1 (50 images) was used consisting of normal controls. These three data sets were taken from the Digital Database for Screening Mammography. For sensitivities of 0.9, 0.85, 0.80, and 0.75, the best results on cancer images exhibit an False-Positive per Image (FPpI) equal to 0.6, 0.45, 0.35, and 0.3, respectively, using a Bayes Linear Discriminant Analysis (LDA) classifier and an FPpI of 0.85, 0.7, 0.55, and 0.45 using a fuzzy inference system (FIS) for false-positive reduction. When the algorithm is tested on normal images, an FPpI equal to 0.4, 0.3, 0.25, and 0.2 was observed using LDA and 0.3, 0.25, 0.2, and 0.15 using the FIS. A preclinical study of an automatic breast mass detection algorithm provided promising results in terms of sensitivity and low false-positive rate. Further development and clinical testing are justified based on the results.

Needs assessment for next generation computer-aided mammography reference image databases and evaluation studies

Breast cancer is globally a major threat for women's health. Screening and adequate follow-up can significantly reduce the mortality from breast cancer. Human second reading of screening mammograms can increase breast cancer detection rates, whereas this has not been proven for current computer-aided detection systems as "second reader". Critical factors include the detection accuracy of the systems and the screening experience and training of the radiologist with the system. When assessing the performance of systems and system components, the choice of evaluation methods is particularly critical. Core assets herein are reference image databases and statistical methods. We have analyzed characteristics and usage of the currently largest publicly available mammography database, the Digital Database for Screening Mammography (DDSM) from the University of South Florida, in literature indexed in Medline, IEEE Xplore, SpringerLink, and SPIE, with respect to type of computer-aided diagnosis (CAD) (detection, CADe, or diagnostics, CADx), selection of database subsets, choice of evaluation method, and quality of descriptions. 59 publications presenting 106 evaluation studies met our selection criteria. In 54 studies (50.9%), the selection of test items (cases, images, regions of interest) extracted from the DDSM was not reproducible. Only 2 CADx studies, not any CADe studies, used the entire DDSM. The number of test items varies from 100 to 6000. Different statistical evaluation methods are chosen. Most common are train/test (34.9% of the studies), leave-one-out (23.6%), and N-fold cross-validation (18.9%). Database-related terminology tends to be imprecise or ambiguous, especially regarding the term "case". Overall, both the use of the DDSM as data source for evaluation of mammography CAD systems, and the application of statistical evaluation methods were found highly diverse. Results reported from different studies are therefore hardly comparable. Drawbacks of the DDSM (e.g. varying quality of lesion annotations) may contribute to the reasons. But larger bias seems to be caused by authors' own decisions upon study design. RECOMMENDATIONS/CONCLUSION: For future evaluation studies, we derive a set of 13 recommendations concerning the construction and usage of a test database, as well as the application of statistical evaluation methods.

“CADEAT”: considerations on the use of CAD (computer-aided diagnosis) in mammography

Computer-aided diagnosis (CAD) has been extensively reported to increase sensitivity by about 10% when added to a single reading while increasing recall rate by 12%, and its current use can be safely recommended in clinical practice. CAD has been suggested as a possible alternative to conventional double reading in screening. Uncontrolled comparison is consistent and suggests that CAD is comparable to double reading in incremental cancer detection rate (CAD +10.6%, double reading +9.1%) and possibly better in recall rate (CAD +12.5%, double reading +28.8%). However, controlled studies comparing single reading + CAD to conventional double reading are not consistent and on average suggest a lower cancer detection rate (-5.1%) and a lower recall rate (-9.8%) for CAD. Scientific evidence is not sufficient for a safe recommendation of single reading + CAD as a current alternative to conventional double reading.

Exploiting E-Infrastructures for Medical Image Storage and Analysis: A Grid Application for Mammography CAD

This paper presents a Digital Repositories Infrastructure (DRI) software platform that simplifies the management of digitalized content and metadata stored on Grid, exploiting its features such as strong security contexts, data federation and large storage and computing capacities. The DRI proposed here includes repository browsing tools, a custom made viewer and a mammograms analysis graphical interface. In general, DRI offers an easy and intuitive interaction model with content stored on the Grid and it also constitutes the foundation to build specific domain applications. In this context, we developed an experimental Grid application in Mammography Computer-Aided Diagnosis (CAD) including a general framework that supports all its stages (image processing, segmentation, training, classification, etc) allowing semiautomatic classification of digital mammograms.

Computer-Aided Diagnosis in Mammography Using Content-based Image Retrieval Approaches: Current Status and Future Perspectives.

As the rapid advance of digital imaging technologies, the content-based image retrieval (CBIR) has became one of the most vivid research areas in computer vision. In the last several years, developing computer-aided detection and/or diagnosis (CAD) schemes that use CBIR to search for the clinically relevant and visually similar medical images (or regions) depicting suspicious lesions has also been attracting research interest. CBIR-based CAD schemes have potential to provide radiologists with "visual aid" and increase their confidence in accepting CAD-cued results in the decision making. The CAD performance and reliability depends on a number of factors including the optimization of lesion segmentation, feature selection, reference database size, computational efficiency, and relationship between the clinical relevance and visual similarity of the CAD results. By presenting and comparing a number of approaches commonly used in previous studies, this article identifies and discusses the optimal approaches in developing CBIR-based CAD schemes and assessing their performance. Although preliminary studies have suggested that using CBIR-based CAD schemes might improve radiologists' performance and/or increase their confidence in the decision making, this technology is still in the early development stage. Much research work is needed before the CBIR-based CAD schemes can be accepted in the clinical practice.

Eigendetection of masses considering false positive reduction and breast density information

The purpose of this article is to present a novel algorithm for the detection of masses in mammographic computer-aided diagnosis systems. Four key points provide the novelty of our approach: (1) the use of eigenanalysis for describing variation in mass shape and size; (2) a Bayesian detection methodology providing a mathematical sound framework, flexible enough to include additional information; (3) the use of a two-dimensional principal components analysis approach to facilitate false positive reduction; and (4) the incorporation of breast density information, a parameter correlated with the performance of most mass detection algorithms and which is not considered in existing approaches. To study the performance of the system two experiments were carried out. The first is related to the ability of the system to detect masses, and thus, free-response receiver operating characteristic analysis was used, showing that the method is able to give high accuracy at a high specificity (80% detection at 1.40 false positives per image). Second, the ability of the system to highlight the pixels belonging to a mass is studied using receiver operating characteristic analysis, resulting in A(z) = 0.89 +/- 0.04. In addition, the robustness of the approach is demonstrated in an experiment where we used the Digital Database for Screening Mammography database for training and the Mammographic Image Analysis Society database for testing the algorithm.

TU‐C‐L100F‐01: New Issues in Breast Imaging

TU‐C‐L100F‐01: New Issues in Breast Imaging

Current status and future directions of computer-aided diagnosis in mammography

The concept of computer-aided detection (CADe) was introduced more than 50 years ago; however, only in the last 20 years there have been serious and successful attempts at developing CADe for mammography. CADe schemes have high sensitivity, but poor specificity compared to radiologists. CADe has been shown to help radiologists find more cancers both in observer studies and in clinical evaluations. Clinically, CADe increases the number of cancers detected by approximately 10%, which is comparable to double reading by two radiologists.

A Microcalcification Detection Using Adaptive Contrast Enhancement on Wavelet Transform and Neural Network

Microcalcification detection is an important part of early breast cancer detection. In this paper, we propose a microcalcification detection algorithm using adaptive contrast enhancement in a mammography CAD (computer-aided diagnosis) system. The proposed microcalcification detection algorithm includes two parts. One is adaptive contrast enhancement in which the enhancement filtering parameters are determined based on noise characteristics of the mammogram. The other is a multi-stage microcalcification detection. The results show that the proposed microcalcification detection algorithm is much more robust against fluctuating noisy environments.

Computerized evaluation of mammographic lesions: what diagnostic role does the shape of the individual microcalcifications play compared with the geometry of the cluster?

The objective of this study was to compare the diagnostic role of features reflecting the geometry of clusters with features reflecting the shape of the individual microcalcification in a mammographic computer-aided diagnosis system. Three hundred twenty-four cases of clustered microcalcifications with biopsy-proven results were digitized at 42-microm resolution and analyzed on a computerized system. The shape factor and number of neighbors were computed for each microcalcification, and the eccentricity of the cluster was computed as well. The shape factor is related to the individual microcalcification; the average number of neighbors and the cluster eccentricity reflect the cluster geometry. Stepwise discriminant analysis was used to evaluate the contribution of the extracted features in predicting malignancy. The performance of a classifier based on the features selected by stepwise discriminant analysis was evaluated by receiver operating characteristic (ROC) analysis. To obtain the best discrimination model, we used stepwise discriminant analysis to select the average number of neighbors and the shape of the individual microcalcification, but excluded cluster eccentricity. A classification scheme assigned the average number of neighbors a weighting factor, which was 1.49 times greater than that assigned to the shape factor of the individual microcalcification. A scheme based only on these two features yielded an ROC curve with an area under the curve (A(z)) of 0.87, indicating a positive predictive value of 61% for 98% sensitivity. Computerized analysis permitted calculations reflecting the shape of individual microcalcification and the geometry of clusters of microcalcifications. For the computerized classification scheme studied, the cluster geometry was more effective in differentiating benign from malignant clusters than was the shape of individual microcalcification.

Computer-Aided Diagnosis in Mammography

Computer-Aided Diagnosis in Mammography

乳がんの先端画像診断−コンピュータ支援診断（CAD）の現状と将来−

Computer-aided diagnosis (CAD) is a diagnosis made by a physician who takes into account the results of quantitative analysis of medical images by computer. The CAD systems, especially called as computer-aided detection (CAD) systems, are now commercially available at least in the two diagnostic fields, i.e., breast cancer diagnosis by mammograms (automated detections of masses and clustered microcalcifications) and thoracic cancer diagnosis by chest radiograms and X-ray CT images (automated nodule detection). Recent studies have indicated promising results in that the computer output can improve the performance of radiologists in detecting lesions on radiographic images. The CAD will make rapid progress to different modalities and various diagnostic regions. In this article, the latest information on mammography CAD for breast image diagnosis is mainly provided with a brief description on ultrasound CAD.

Performance Change of Mammographic CAD Schemes Optimized with Most-Recent and Prior Image Databases

Rationale and Objectives The authors evaluated performance changes in the detection of masses on “current” (latest) and “prior” images by computer-aided diagnosis (CAD) schemes that had been optimized with databases of current and prior mammograms. Materials and Methods The authors selected 260 pairs of matched consecutive mammograms. Each current image depicted one or two verified masses. All prior images had been interpreted originally as negative or probably benign. A CAD scheme initially detected 261 mass regions and 465 false-positive regions on the current images, and 252 corresponding mass regions (early signs) and 471 false-positive regions on prior images. These regions were divided into two training and two testing databases. The current and prior training databases were used to optimize two CAD schemes with a genetic algorithm. These schemes were evaluated with two independent testing databases. Results The scheme optimized with current images produced areas under the receiver operating characteristic curve of 0.89 ± 0.01 and 0.65 ± 0.02 when tested with current images and prior images, respectively. The scheme optimized with prior images produced areas under the receiver operating characteristic curve of 0.81 ± 0.02 and 0.71 ± 0.02 when tested with current images and prior images, respectively. Performance changes for both current and prior testing databases were significant ( P < .01) for the two schemes. Conclusion CAD schemes trained with current images do not perform optimally in detecting masses depicted on prior images. To optimize CAD schemes for early detection, it may be important to include in the training database a large fraction of prior images originally reported as negative and later proven to be positive.

Computer-aided diagnosis of breast cancer in mammography: evidence and potential.

Computer-aided diagnosis (CAD) methods are being developed to help radiologists improve the interpretation of mammograms for the detection of breast cancer. We review several laboratory observer performance studies of computer-aided diagnosis of malignant and benign breast lesions. These studies show that CAD can improve radiologist diagnostic performance by increasing the number of their biopsy recommendations for actual malignant lesions while reducing the number of their biopsy recommendations for suspicious but actually benign lesions, and by reducing the variability in their interpretation of mammograms. These results indicate a potential clinical role of CAD in mammography for the detection of breast cancer.

Learning contextual relationships in mammograms using a hierarchical pyramid neural network.

This paper describes a pattern recognition architecture, which we term hierarchical pyramid/neural network (HPNN), that learns to exploit image structure at multiple resolutions for detecting clinically significant features in digital/digitized mammograms. The HPNN architecture consists of a hierarchy of neural networks, each network receiving feature inputs at a given scale as well as features constructed by networks lower in the hierarchy. Networks are trained using a novel error function for the supervised learning of image search/detection tasks when the position of the objects to be found is uncertain or ill defined. We have evaluated the HPNN's ability to eliminate false positive (FP) regions of interest generated by the University of Chicago's (UofC) Computer-aided diagnosis (CAD) systems for microcalcification and mass detection. Results show that the HPNN architecture, trained using the uncertain object position (UOP) error function, reduces the FP rate of a mammographic CAD system by approximately 50% without significant loss in sensitivity. Investigation into the types of FPs that the HPNN eliminates suggests that the pattern recognizer is automatically learning and exploiting contextual information. Clinical utility is demonstrated through the evaluation of an integrated system in a clinical reader study. We conclude that the HPNN architecture learns contextual relationships between features at multiple scales and integrates these features for detecting microcalcifications and breast masses.