Kevin S. Woods

COMPARATIVE EVALUATION OF PATTERN RECOGNITION TECHNIQUES FOR DETECTION OF MICROCALCIFICATIONS IN MAMMOGRAPHY

Computer-assisted detection of microcalcifications in mammographic images will likely require a multistage algorithm that includes segmentation of possible microcalcifications, pattern recognition techniques to classify the segmented objects, a method to determine if a cluster of calcifications exists, and possibly a method to determine the probability of malignancy. This paper focuses on the first three of these stages, and especially on the classification of segmented local bright spots as either calcification or noncalcification. Seven classifiers (linear and quadratic classifiers, binary decision trees, a standard backpropagation network, 2 dynamic neural networks, and a K-nearest neighbor) are compared. In addition, a postprocessing step is performed on objects identified as calcifications by the classifiers to determine if any clusters of microcalcifications exist. A database of digitized film mammograms is used for training and testing. Detection accuracy of individual and clustered microcalcificat...

Combination of multiple classifiers using local accuracy estimates

Combination of multiple classifiers (CMC) has recently drawn attention as a method of improving classification accuracy. This paper presents a method for combining classifiers that use estimates of each individual classifiers local accuracy in small regions of feature space surrounding an unknown test sample. Only the output of the most locally accurate classifier is considered. We address issues of (1) optimization of individual classifiers, and (2) the effect of varying the sensitivity of the individual classifiers on the CMC algorithm. Our algorithm performs better on data from a real problem in mammogram image analysis than do other recently proposed CMC techniques.

Image segmentation in digital mammography: comparison of local thresholding and region growing algorithms.

Local thresholding and region-growing algorithms are developed and applied to digitized mammograms to quantify the parenchymal densities. The algorithms are first evaluated and optimized on phantom images reflecting varying image contrast, X-ray exposure conditions, and time-related changes. The difference between the segmentation results of the two techniques is less than 6% on the phantom images and 11% on the mammograms. The agreement between the computerized procedures and a manual one is in the range of 74-98%, depending on the breast parenchymal pattern and segmentation algorithm. The results show that computerized parenchymal classification of digitized mammograms is possible and independent of exposure.

The detection of micro-calcifications in mammographic images using high dimensional features

This paper examines techniques for the efficient use of high dimensional feature sets in the detection of micro-calcifications in mammograms. The paper focuses on techniques for dimensionality reduction and discriminant analysis. The paper examines the use of principal components and Fishers linear discriminant for dimensionality reduction along with parametric and nonparametric statistical techniques for discriminant analysis.<<ETX>>

A neural network approach to microcalcification detection

A supervised dynamic neural network is used to detect microcalcifications in digitized mammograms. A segmentation process is used to extract candidate objects from the mammogram, and then the neural network is used to determine if the candidate object is a microcalcification. A simple postprocessing procedure is applied to the results to check for clusters of microcalcifications. The neural network method is compared to the K-nearest neighbor method. The artificial neural network (ANN) used for pattern classification is called cascade correlation (CC). The true positive detection rate of the CC ANN for individual microcalcifications is 73% and 92% for nonmicrocalcifications. >

Enhancement of Digitized Mammograms Using a Local Thresholding Technique

In this paper, a local threshdding technique is used to improve the contrast of dighzed mammograms. Dighzed mammograms cannot be segmented using a sngle threshold value due to their lack of contrast. There is also a degredatm of image quality, speafdly in conlrast, when a mammogram is dgitrzed. The technique presented here deals wdh these problems by calculaling a threshold value for each pixel in the image based on the intensibes of the neighbonng puels in a wndow. The contrast of the image is improved by darkening those pxels which fall bebw the threshold After an image is processed uang this algonthm, the resubng image can be easily segmented using a single threshdd value. Th~s method could also be used on other images wdh smilar characterisbcs as those of the digibzed mammograms.

Application of nonlinear filtering in mammograms

A computer assisted method for the quantification and classification of mammographic parenchymal patterns (MPP) is proposed. Enhancement of mammographic images is performed using order statistic filtering, a superior method compared to median filtering techniques previously reported. Two complementary methods are proposed for quantification and classification of MPP, a local thresholding technique and an edge detection method, respectively. The latter method is based on non-linear filtering which uses order statistics or linear combination of order statistics filter specifically tailored to identify the boundaries and fine details of MPP. The edge detection method proved to be useful for the difficult differentiation of Wolfes P2 and DY MPP that have similar breast density and common characteristics. The results suggest that the methods proposed are potentially useful for identification and quantitation of MPPs as required for mass screening of breast cancer.