Authors : R.D. Ghongade and D.G. Wakde
Abstract: Neural network is utilized as rising diagnosis tool for cancer disease. The goal of this exploration is to determine tumor growth in breast with a machine learning method based on RF, ELM and RF-ELM classifier. MIAS database is utilized for the advanced mammogram images. Pre-processing is for the most part expected to enhance the low nature of image. The ROI is resolved by the measure of suspicious region. After the suspicious area is portioned, features are extracted by texture analysis. GLCM is utilized as a surface credit to extricate the suspicious region. From all extracted features best features are chosen with the assistance of CBF and PCA. RF, ELM and RF-ELM are utilized as classifier. The consequences of present resarch demonstrate that the CAD framework utilizing RF-ELM classifier is exceptionally compelling and accomplishes the best outcome in the finding of breast malignancy.
R.D. Ghongade and D.G. Wakde, 2017. Random Forest and Extreme Learning Machine Based CAD System for Breast Cancer. Journal of Engineering and Applied Sciences, 12: 9477-9482.