Feature Selection based on Genetic Algorithm for Classification of Mammogram Using K-means, k-NN and Euclidean Distance
DOI:
https://doi.org/10.25130/tjps.v22i9.883Abstract
There have been several supervised classification attempts for mammograms in the recent times, but very few research works have focused on unsupervised classification to explore its potentialities and weaknesses. I have in this paper attempted to utilize unsupervised clusters to classify malignant, and benign mammograms samples. MiniMIAS database has total 322 mammogram images out which 64 are benign and 51 are malignant. I used 115 images for my experimentation i.e. 64 benign and 51 malignant. Out of these 115, 60% were used for training and 40% for testing. Therefore from 64 benign cases 39 images were used for training and rest for testing, and out of 51 malignant cases 31 images were used for training and rest for testing., the classifications was done on the bases of the features selected using genetic algorithm. Attempts have also been made to study the performance of each feature selected by Genetic Algorithm (GA) in classification. The initially identified clusters using K-means are used to classify 60 unknown samples using k-NN. The proposed work got reasonably good results with 96.23% accuracy for malignant samples, 95.37% for benign. The proposed work can help the radiologists and oncologist as second opinion during screening sessions for early detection.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Tikrit Journal of Pure Science
This work is licensed under a Creative Commons Attribution 4.0 International License.
Tikrit Journal of Pure Science is licensed under the Creative Commons Attribution 4.0 International License, which allows users to copy, create extracts, abstracts, and new works from the article, alter and revise the article, and make commercial use of the article (including reuse and/or resale of the article by commercial entities), provided the user gives appropriate credit (with a link to the formal publication through the relevant DOI), provides a link to the license, indicates if changes were made, and the licensor is not represented as endorsing the use made of the work. The authors hold the copyright for their published work on the Tikrit J. Pure Sci. website, while Tikrit J. Pure Sci. is responsible for appreciate citation of their work, which is released under CC-BY-4.0, enabling the unrestricted use, distribution, and reproduction of an article in any medium, provided that the original work is properly cited.