This paper presents classification of brain cancer using wavelet transformation and Adaptive Neighborhood Based Modified Backpropagation (ANMBP). Three stages of the processes, namely features extraction, features reduction, and classification process. Wavelet transformation is used for feature extraction and ANMBP is used for classification process. The result of features extraction is feature vectors. Features reduction used 100 energy values per feature and 10 energy values per feature. Classifications of brain cancer are normal, alzheimer, glioma, and carcinoma. Based on simulation results, 10 energy values per feature can be used to classify brain cancer correctly. The correct classification rate of proposed system is 95 %. This research demonstrated that wavelet transformation can be used for features extraction and ANMBP can be used for classification of brain cancer.
Application of wavelet transformation and adaptive neighborhood based modified backpropagation (ANMBP) for classification of brain cancer
Indah Werdiningsih, Badrus Zaman, Barry Nuqoba; Application of wavelet transformation and adaptive neighborhood based modified backpropagation (ANMBP) for classification of brain cancer. AIP Conf. Proc. 1 August 2017; 1867 (1): 020004. https://doi.org/10.1063/1.4994407
Download citation file: