Breast Cancer Detection Using KNN Classifier
  • Author(s): SANTHANA LAKSHMI.N ; CHITRA EVANGELIN CHRISTINA.M
  • Paper ID: 1701134
  • Page: 76-79
  • Published Date: 20-04-2019
  • Published In: Iconic Research And Engineering Journals
  • Publisher: IRE Journals
  • e-ISSN: 2456-8880
  • Volume/Issue: Volume 2 Issue 10 April-2019
Abstract

Diseases are analyzed by different digital image processing techniques. Early detection of breast cancer can improve survival rates to a great extent. Inter-observer and intra-observer errors occur frequently in analysis of medical images, given the high variability between interpretations of different radiologists. Breast cancer detection involves the steps which includes breast image preprocessing, tumor detection, feature extraction, training data generation, and classifier training. In the breast image preprocessing, denoising and enhancing contrast processes on the original mammogram have been utilized to increase the contrast between the masses and the surrounding tissues. The tumor detection is then performed to localize the tumor ROI. After that, features including morphological features, texture features and density features are extracted from the detected ROI. During the training process, the KNN (K Nearest Neighbour) classifier have been trained with every image from the breast image dataset using their extracted features and corresponding labels. GMM(Gaussian Mixture Model) segmentation used which enhances the contrast of the image to improve its visual quality.The final output in this project finally describes the presence of breast cancer for the given input data.

Keywords

Breast cancer, GMM (Gaussian Mixture Model), KNN (K Neareast Neighbour)

Citations

IRE Journals:
SANTHANA LAKSHMI.N , CHITRA EVANGELIN CHRISTINA.M "Breast Cancer Detection Using KNN Classifier" Iconic Research And Engineering Journals Volume 2 Issue 10 2019 Page 76-79

IEEE:
SANTHANA LAKSHMI.N , CHITRA EVANGELIN CHRISTINA.M "Breast Cancer Detection Using KNN Classifier" Iconic Research And Engineering Journals, 2(10)