Rangaraj M. Rangayyan

Computer-Aided Detection and Diagnosis of Breast Cancer With Mammography: Recent Advances

Breast cancer is the second-most common and leading cause of cancer death among women. It has become a major health issue in the world over the past 50 years, and its incidence has increased in recent years. Early detection is an effective way to diagnose and manage breast cancer. Computer-aided detection or diagnosis (CAD) systems can play a key role in the early detection of breast cancer and can reduce the death rate among women with breast cancer. The purpose of this paper is to provide an overview of recent advances in the development of CAD systems and related techniques. We begin with a brief introduction to some basic concepts related to breast cancer detection and diagnosis. We then focus on key CAD techniques developed recently for breast cancer, including detection of calcifications, detection of masses, detection of architectural distortion, detection of bilateral asymmetry, image enhancement, and image retrieval.

Region-based contrast enhancement of mammograms

Diagnostic features in mammograms vary widely in size and shape. Classical image enhancement techniques cannot adapt to the varying characteristics of such features. An adaptive method for enhancing the contrast of mammographic features of varying size and shape is presented. The method uses each pixel in the image as a seed to grow a region. The extent and shape of the region adapt to local image gray-level variations, corresponding to an image feature. The contrast of each region is calculated with respect to its individual background. Contrast is then enhanced by applying an empirical transformation based on each regions seed pixel value, its contrast, and its background. A quantitative measure of image contrast improvement is also defined based on a histogram of region contrast and used for comparison of results. Using mammogram images digitized at high resolution (less than 0.1 mm pixel size), it is shown that the validity of microcalcification clusters and anatomic details is considerably improved in the processed images.

Application of shape analysis to mammographic calcifications

The authors have developed a set of shape factors to measure the roughness of contours of calcifications in mammograms and for use in their classification as malignant or benign. The analysis of mammograms is performed in three stages. First, a region growing technique is used to obtain the contours of calcifications. Then, three measures of shape features, including compactness, moments, and Fourier descriptors are computed for each region. Finally, their applicability for classification is studied by using the three shape measures to form feature vectors. Classification of 143 calcifications from 18 biopsy-proven cases as benign or malignant using the three measures with the nearest-neighbor method was 100% accurate.

Measures of acutance and shape for classification of breast tumors

Most benign breast tumors possess well-defined, sharp boundaries that delineate them from surrounding tissues, as opposed to malignant tumors. Computer techniques proposed to date for tumor analysis have concentrated on shape factors of tumor regions and texture measures. While shape measures based on contours of tumor regions can indicate differences in shape complexities between circumscribed and spiculated tumors, they are not designed to characterize the density variations across the boundary of a tumor. Here, the authors propose a region-based measure of image edge profile acutance which characterizes the transition in density of a region of interest (ROI) along normals to the ROI at every boundary pixel. The authors investigate the potential of acutance in quantifying the sharpness of the boundaries of tumors, and propose its application to discriminate between benign and malignant mammographic tumors. In addition, they study the complementary use of various shape factors based upon the shape of the ROI, such as compactness. Fourier descriptors, moments, and chord-length statistics to distinguish between circumscribed and spiculated tumors. Thirty-nine images from the Mammographic Image Analysis Society (MIAS) database and an additional set of 15 local cases were selected for this study. The cases included 16 circumscribed benign, 7 circumscribed malignant, 12 spiculated benign, and 19 spiculated malignant lesions. All diagnoses were proven by pathologic examinations of resected tissue. The contours of the lesions were first marked by an expert radiologist using X-Paint and X-Windows on a SUN-SPARCstation 2 Workstation. For computation of acutance, the ROI boundaries were iteratively approximated using a split/merge and end-point adjustment technique to obtain the best-fitting polygonal approximation. The jackknife method using the Mahalanobis distance measure in the BMDP (Biomedical Programs) package was used for classification of the lesions using acutance and the shape factors as features in various combinations. Acutance alone resulted in a benign/malignant classification accuracy of 95% the MIAS cases. Compactness alone gave a circumscribed/spiculated classification rate of 92.3% with the MIAS cases. Acutance in combination with a moment-based shape measure and a Fourier descriptor-based measure gave four-group classification rate of 95% with the MIAS cases. The results indicate the importance of including lesion edge definition with shape information for classification of tumors, and that the proposed measure of acutance fills this need.

A review of computer-aided diagnosis of breast cancer : Toward the detection of subtle signs

Abstract Mammography is the best available tool for screening for the early detection of breast cancer. Mammographic screening has been shown to be effective in reducing breast cancer mortality rates: screening programs have reduced mortality rates by 30–70%. Mammograms are difficult to interpret, especially in the screening context. The sensitivity of screening mammography is affected by image quality and the radiologists level of expertise. Computer-aided diagnosis (CAD) technology can improve the performance of radiologists, by increasing sensitivity to rates comparable to those obtained by double reading, in a cost-effective manner. Current research is directed toward the development of digital imaging and image analysis systems that can detect mammographic features, classify them, and provide visual prompts to the radiologist. Radiologists would like the ability to change the contrast of a mammogram, either manually or with pre-selected settings. Computer techniques for detecting, classifying, and annotating diagnostic features on the images would be desirable. This paper presents an overview of digital image processing and pattern analysis techniques to address several areas in CAD of breast cancer, including: contrast enhancement, detection and analysis of calcifications, detection and analysis of masses and tumors, analysis of bilateral asymmetry, and detection of architectural distortion. Although a few commercial CAD systems have been released, the detection of subtle signs of breast cancer such as global bilateral asymmetry and focal architectural distortion remains a difficult problem. We present some of our recent works on the development of image processing and pattern analysis techniques for these applications.

Gradient and texture analysis for the classification of mammographic masses

Computer-aided classification of benign and malignant masses on mammograms is attempted in this study by computing gradient-based and texture-based features. Features computed based on gray-level co-occurrence matrices (GCMs) are used to evaluate the effectiveness of textural information possessed by mass regions in comparison with the textural information present in mass margins. A method involving polygonal modeling of boundaries is proposed for the extraction of a ribbon of pixels across mass margins. Two gradient-based features are developed to estimate the sharpness of mass boundaries in the ribbons of pixels extracted from their margins. A total of 54 images (28 benign and 26 malignant) containing 39 images from the Mammographic Image Analysis Society (MIAS) database and 15 images from a local database are analyzed. The best benign versus malignant classification of 82.1%, with an area (A/sub z/) of 0.85 under the receiver operating characteristics (ROC) curve, was obtained with the images from the MIAS database by using GCM-based texture features computed from mass margins. The classification method used is based on posterior probabilities computed from Mahalanobis distances. The corresponding accuracy using jack-knife classification was observed to be 74.4%, with A/sub x/=0.67. Gradient-based features achieved A/sub x/=0.6 on the MIAS database and A/sub z/=0.76 on the combined database. The corresponding values obtained using jack-knife classification were observed to be 0.52 and 0.73 for the MIAS and combined databases, respectively.

Detection of breast masses in mammograms by density slicing and texture flow-field analysis

We propose a method for the detection of masses in mammographic images that employs Gaussian smoothing and subsampling operations as preprocessing steps. The mass portions are segmented by establishing intensity links from the central portions of masses into the surrounding areas. We introduce methods for analyzing oriented flow-like textural information in mammograms. Features based on flow orientation in adaptive ribbons of pixels across the margins of masses are proposed to classify the regions detected as true mass regions or false-positives (FPs). The methods yielded a mass versus normal tissue classification accuracy represented as an area (A/sub z/) of 0.87 under the receiver operating characteristics (ROCs) curve with a dataset of 56 images including 30 benign disease, 13 malignant disease, and 13 normal cases selected from the mini Mammographic Image Analysis Society database. A sensitivity of 81% was achieved at 2.2 FPs/image. Malignant tumor versus normal tissue classification resulted in a higher A/sub z/ value of 0.9 under the ROC curve using only the 13 malignant and 13 normal cases with a sensitivity of 85% at 2.45 FPs/image. The mass detection algorithm could detect all the 13 malignant tumors successfully, but achieved a success rate of only 63% (19/30) in detecting the benign masses. The mass regions that were successfully segmented were further classified as benign or malignant disease by computing five texture features based on gray-level co-occurrence matrices (GCMs) and using the features in a logistic regression method. The features were computed using adaptive ribbons of pixels across the boundaries of the masses. Benign versus malignant classification using the GCM-based texture features resulted in A/sub z/=0.79 with 19 benign and 13 malignant cases.

Feature enhancement of film mammograms using fixed and adaptive neighborhoods

Digital techniques are presented for xerographylike enhancement of features in film mammograms. The mammographic image is first digitized using a procedure for gray scale dynamic range expansion. A pixel operator is then applied to the image, which performs contrast enhancement according to a specified function. The final transformation leads to either a positive or negative mode display as desired. We also present an adaptive neighborhood feature enhancement technique that enhances visibility of objects and details in an image. The availability of the enhanced images should aid diagnosis of breast cancer without requiring additional x-ray dose such as for xeromammography.

Automatic identification of the pectoral muscle in mammograms

The pectoral muscle represents a predominant density region in most medio-lateral oblique (MLO) views of mammograms; its inclusion can affect the results of intensity-based image processing methods or bias procedures in the detection of breast cancer. Local analysis of the pectoral muscle may be used to identify the presence of abnormal axillary lymph nodes, which may be the only manifestation of occult breast carcinoma. We propose a new method for the identification of the pectoral muscle in MLO mammograms based upon a multiresolution technique using Gabor wavelets. This new method overcomes the limitation of the straight-line representation considered in our initial investigation using the Hough transform. The method starts by convolving a group of Gabor filters, specially designed for enhancing the pectoral muscle edge, with the region of interest containing the pectoral muscle. After computing the magnitude and phase images using a vector-summation procedure, the magnitude value of each pixel is propagated in the direction of the phase. The resulting image is then used to detect the relevant edges. Finally, a post-processing stage is used to find the true pectoral muscle edge. The method was applied to 84 MLO mammograms from the Mini-MIAS (Mammographic Image Analysis Society, London, U.K.) database. Evaluation of the pectoral muscle edge detected in the mammograms was performed based upon the percentage of false-positive (FP) and false-negative (FN) pixels determined by comparison between the numbers of pixels enclosed in the regions delimited by the edges identified by a radiologist and by the proposed method. The average FP and FN rates were, respectively, 0.58% and 5.77%. Furthermore, the results of the Gabor-filter-based method indicated low Hausdorff distances with respect to the hand-drawn pectoral muscle edges, with the mean and standard deviation being 3.84/spl plusmn/1.73 mm over 84 images.

Boundary modelling and shape analysis methods for classification of mammographic masses.

The problem of computer-aided classification of benign and malignant breast masses using shape features is addressed. The aim of the study is to look at the exceptions in shapes of masses such as circumscribed malignant tumours and spiculated benign masses which are difficult to classify correctly using common shape analysis methods. The proposed methods of shape analysis treat the objects boundary in terms of local details. The boundaries of masses analysed using the proposed methods were manually drawn on mammographic images by an expert radiologist (JELD). A boundary segmentation method is used to separate major portions of the boundary and to label them as concave or convex segments. To analyse the shape information localised in each segment, features are computed through an iterative procedure for polygonal modelling of the mass boundaries. Features are based on the concavity fraction of a mass boundary and the degree of narrowness of spicules as characterised by a spiculation index. Two features comprising spiculation index (SI) and fractional concavity (fcc) developed in the present study when used in combination with the global shape feature of compactness resulted in a benign/malignant classification accuracy of 82%, with an area (Az) of 0.79 under the receiver operating characteristics (ROC) curve with a database of the boundaries of 28 benign masses and 26 malignant tumours. SI alone resulted in a classification accuracy of 80% with Az of 0.82. The combination of all the three features achieved 91% accuracy of circumscribed versus spiculated classification of masses based on shape.