Paper Title
Tree Encoder Based Mammogram Image Compression And Its Application To Classification Of Microcalcification

Medical image processing is a backbone for the rapid development in the field of medical science. Huge diagnosis in the medical field depends on image processing which is due to the rapid development in computer aided diagnosis. This article proposes a tree encoder based mammogram image compression and its application to classification of micro calcification. Mammogram images are huge in size which requires preprocessing, compression and classification process for diagnosis. In this work, mammogram images are compressed using tree based encoder SPIHT in order to reduce the size of image. Further the image is preprocessed to remove the redundant portion using improved watershed algorithm. Then the image is classified as benign and malignant using training data set. The experiment has been carried out on the various mammogram images obtained from MIAS database. The results seem to be promising in terms of compression ratio and PSNR for compression and in terms of accuracy, sensitivity and specificity for classification