High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Iterative Quad Tree Decomposition Based Selection, and Detection and Segmentation of Breast Tumor Images using Region Growing Snake Contours

Author(s):

Kulveer Kaur , Swami Vivekanand Institutes of Engineering & Technology Banur (Chandigarh); Ms. Manmeen Kaur, Swami Vivekanand Institutes of Engineering & Technology Banur (Chandigarh)

Keywords:

Snake Contour Segmentation, QTD Algorithm, Ultrasound Images, Tumor Localization

Abstract

Mammography is currently the best method for detecting a breast cancer early, before the malignant tissue is substantial enough to feel or cause symptoms. However, the interpretation of a mammogram is often difficult and depends on the expertise and experience of the radiologist. Most of the problems or limitations in mammography can be overcome by using digital image processing techniques. Computer-aided-diagnosis (CAD) system can be used to assist the radiologists and the physicians analyses the overall images, and find tumors that a radiologist might not spot. Combining computer-aided-diagnosis with mammography will improve the ability to find cancer. There are number methods researched by scholars for medical image segmentation in which region growing and contour gradient based methods work efficiently at the tumor boundaries. But major requirement for such methods need a starting point or called as seed point in between the tumor. Their fore accuracy of actual tumor detection depends highly on the truly detected seed point.. Also the second drawback is their location such that they found near the boundary areas of tumor which results in the leakage of segmentation algorithms through weak boundary points. All these limitations has been covered in this work. Iterative quad tree has been improved with the help of morphological operations which gives seed point at the centroid of the tumor. Experimental results shows that proposed method gives high accuracy in truly segmentation of breast tumors in collected dataset.

Other Details

Paper ID: IJSRDV6I110197
Published in: Volume : 6, Issue : 11
Publication Date: 01/11/2019
Page(s): 337-340

Article Preview

Download Article