Computer-Aided Diagnosis of Malignant Mammograms using Zernike Moments and SVM

Abstract

This work is directed toward the development of a computer-aided diagnosis (CAD) system to detect abnormalities or suspicious areas in digital mammograms and classify them as malignant or nonmalignant. Original mammogram is preprocessed to separate the breast region from its background. To work on the suspicious area of the breast, region of interest (ROI) patches of a fixed size of 128 × 128 are extracted from the original large-sized digital mammograms. For training, patches are extracted manually from a preprocessed mammogram. For testing, patches are extracted from a highly dense area identified by clustering technique. For all extracted patches corresponding to a mammogram, Zernike moments of different orders are computed and stored as a feature vector. A support vector machine (SVM) is used to classify extracted ROI patches. The experimental study shows that the use of Zernike moments with order 20 and SVM classifier gives better results among other studies. The proposed system is tested on Image Retrieval In Medical Application (IRMA) reference dataset and Digital Database for Screening Mammography (DDSM) mammogram database. On IRMA reference dataset, it attains 99 % sensitivity and 99 % specificity, and on DDSM mammogram database, it obtained 97 % sensitivity and 96 % specificity. To verify the applicability of Zernike moments as a fitting texture descriptor, the performance of the proposed CAD system is compared with the S. Sharma · P. Khanna ( ) Pandit Dwarka Prasad Mishra Indian Institute of Information Technology, Design and Manufacturing Jabalpur, Dumna Airport Road, P.O.: Khamaria, Jabalpur, Madhya Pradesh 482 005, India e-mail: pkhanna@iiitdmj.ac.in S. Sharma e-mail: shubhi.sharma@iiitdmj.ac.in other well-known texture descriptors namely gray-level cooccurrence matrix (GLCM) and discrete cosine transform (DCT).

Classification of Mammograms using Various Feature Extraction Methods and Machine Learning

“…In Reference [8] the author developed a CAD system where the suspicious regions were extracted from mammograms and Zernike moments of different order was used to create feature vector and classification of DDSM dataset took place by SVM classifier. Reference [9] presented content based image retrieval (CBIR) system for retrieving images based upon their content.…”

Breast Segmentation and Probable Region Identification for Breast Cancer using DL-CNN

“…Moreover, the parameters had to be modified based on the datasets. Similarly, In [17] the method tries to first extract the patches of 128 X 128 pixels from the pre-processed image for the training and for the testing it is extracted from the dense area, these are performed manually. Hence, CAD [18] was presented for abnormality classification where the ROI (Region of Interest) are pointed by the three radiologists and it is cropped into 128X 128 pixel.…”

Study of wireless mammography image transmission impacts on robust cyber-aided diagnosis systems

“…Therefore the interpretation of wireless mammogram result is difficult, and it may cause a high proportion of unnecessary biopsy, and late treatment of cancer [2] [3] [6]. In order to increase the effectiveness and efficiency of mammography screening procedure, a wireless version of computer-aided diagnosis (CAD) system is introduced as second opinion to assist radiologists to read and interpret images.…”

“…Different pre-processing methods such as Intensity based pre-processing, histogram equalization and wavelet transformation are used to prepare mammography images in the CAD systems [6] [9] [10].…”

“…In order to increase the effectiveness and efficiency of mammography screening procedure, a wireless version of computer-aided diagnosis (CAD) system is introduced as second opinion to assist radiologists to read and interpret images. The combination of the wireless CAD and experts' knowledge can improve the accuracy, specificity and sensitivity and also reduce the image reading time [6] [7].…”