Detection of Microcalcifications in Digital Mammograms using Fuzzy Euler Graph Segmentation method D.Saraswathi Department of ECE, Manakula Vinayagar Institute of Technology, Madagadipet, Puducherry, India saraswathi.mit@gmail.com E.Srinivasan Department of ECE, Pondicherry Engineering College Puducherry, India Abstract— This work presents the Fuzzy Euler graph based segmentation algorithm to segment the microcalcifications present in mammogram images. Mammography is the best method for diagnosing early stages of breast cancer. Computer Aided Diagnosis system lacks accuracy and it is time consuming. Segmentation plays a vital role in detecting the cancerous microcalcifications and in turn increases the classification accuracy. The proposed algorithm identifies the various size and shape of the individual clusters of microcalcifications present in the mammogram image. The sensitivity, the specificity, the false positives and the false negatives of the proposed algorithm are determined. The simulation results proved that the suggested approach has high segmentation accuracy. The originality of this work lies in the segmentation of mammogram image using Fuzzy concept in Euler Graph segmentation method. Keywords- Mammogram; Segmentation; Computer Aided Diagnosis system; Graph Theory; Euler Graphs; Fuzzy theory I. INTRODUCTION Diagnostic imaging is an invaluable tool in medicine. Magnetic resonance imaging (MRI), computed tomography (CT), digital mammography, and other imaging modalities provide an effective means for noninvasively mapping the anatomy of a subject. These technologies have greatly increased knowledge of normal and diseased anatomy for medical research and are a critical component in diagnosis and treatment planning. The growing size and number of these medical images have necessitated the use of computers to facilitate processing and analysis. In particular, computer algorithms for the delineation of anatomical structures and other regions of interest are becoming increasingly important in assisting and automating specific radiological tasks. These algorithms, called image segmentation algorithms, play a vital role in numerous biomedical-imaging applications, such as the quantification of tissue volumes, diagnosis, localization of pathology, study of anatomical structure, treatment planning, and computer-integrated surgery. Several common approaches have appeared in the recent literature on medical image segmentation. Mehmet Sezgin et. al [1] conducted an exhaustive survey which includes 40 image thresholding methods . They categorized image thresholding methods into six categories such as histogram shape-based methods, clustering-based methods, entropy-based methods, object attribute-based methods, the spatial methods and local methods. They have used different performance criteria such as misclassification error, edge mismatch, relative foreground area error, modified Hausdorff distance, and region nonuniformity. Daniel Cremers et. al [2] have demonstrated how the level set function is driven to partition the image plane into domains of coherent color, texture, dynamic texture or motion. Dzhung. L . Pham et. al [3] divided the medical image segmentation methods into 8 categories. They are (1) thresholding approaches, (2) region growing approaches, (3) classifiers, (4) clustering approaches, (5) Markov random field models, (6) artificial neural networks, (7) deformable models, and (8) atlas guided approaches. The authors also discussed the advantages and disadvantages of these methods. J. Alison Noble et. al [4] have given a detailed survey for medical B-mode ultrasound images. They mainly focused and reviewed the segmentation of echocardiography, breast ultrasound, transrectal ultrasound, intravascular ultrasound images, and ultrasound images acquired in obstetrics and gynecology. They also summarized the performance assessment and clinical validation. Among different segmentation schemes, graph theoretical ones have several good features in practical applications. It explicitly organizes the image elements into mathematically sound structures, and makes the formulation of the problem more flexible and the computation more efficient. Bo Peng et. al [5] conducted a systematic survey of graph theoretical methods for image segmentation. They categorized the methods into five classes, namely the minimal spanning tree based methods, graph cut based methods with cost functions, graph cut based methods on Markov random field models, the shortest path based methods and some other methods. They carried out the quantitative evaluation by using five indices – Probabilistic Rand index, Normalized Probabilistic Rand index, Variation of Information, Global Consistency Error and Boundary Displacement. Basava prasad B et. al [6] carried out an organized survey and classified the image segmentation methods under 3 categories namely the traditional methods, graph theoretical methods and combination of both traditional and graph theoretical methods, which are flexible, cost effective and computationally more efficient. Pasquale Foggia et. al [7] presented a graph-based clustering method and proved that the proposed method is consistently better than the Vol. 14 CIC 2016 Special Issue International Journal of Computer Science and Information Security (IJCSIS) https://sites.google.com/site/ijcsis/ ISSN 1947-5500 International Conference on Advances in Computational Intelligence and Communication (CIC 2016) Pondicherry Engineering College, Puducherry, India October 19 & 20 - 2016 27