IJENS Publisher Indexed In SCOPUS

Navigation


Title: Detection and Classification of Normal and Abnormal Heart Diseases Using SIFT and SVM
Author(s): M. Balasubramanian, M. Jannathl Firdouse
Pages: 1-8 Paper ID:183901-5757-IJVIPNS-IJENS Published: February, 2018
Abstract: The lung cancer only accounts more deaths than any other type of cancers in both men and women. Lung zones are identified by assessing the lungs by comparing the upper, middle and lower lung parts on the left and right. Asymmetry of lung density is denoted as either abnormal whiteness (increased density), or ab-normal blackness (decreased density). Once the asymmetry is spotted, the next step is to decide which side is abnormal. If there is an area that is different from the surrounding is bilateral lung, then this is likely to be the abnormal area. If the alveoli and small airways fill with dense material, the lung is said to be consolidat-ed. The early detection of lung cancer makes the survival rate of patients. In this paper, we proposed a method for the detection and classification of chest images using Scale Invariant Feature Transform and Support Vector Machine. The SIFT descriptor detects the key points from the gray level images. Seven features are detected by applying the SVM principle with SIFT transform. The performance measures such as precision, recall, accuracy and F-Score are estimated. We have collected more than 100 CT imag-es for this classification from Radiological center, KIMS Hospital. The normal and abnormal chest images are classified with the better performance measures.
Keywords: SIFT, SVM, Normal and abnormal images, Precision, Recall, F-Score
Full Text (.pdf)  International Journals Of Engineering and Science | 1,125 KB