Elm Based Cad System to Classify Mammograms by the Combination of CLBP and Contourlet

Abstract

Breast cancer is a serious life threat to the womanhood, worldwide. Mammography is the promising screening tool, which can show the abnormality being detected. However, the physicians find it difficult to detect the affected regions, as the size of microcalcifications is very small. Hence it would be better, if a CAD system can accompany the physician in detecting the malicious regions. Taking this as a challenge, this paper presents a CAD system for mammogram classification which is proven to be accurate and reliable. The entire work is decomposed into four different stages and the outcome of a phase is passed as the input of the following phase. Initially, the mammogram is pre-processed by adaptive median filter and the segmentation is done by GHFCM. The features are extracted by combining the texture feature descriptors Completed Local Binary Pattern (CLBP) and contourlet to frame the feature sets. In the training phase, Extreme Learning Machine (ELM) is trained with the feature sets. During the testing phase, the ELM can classify between normal, malignant and benign type of cancer. The performance of the proposed approach is analysed by varying the classifier, feature extractors and parameters of the feature extractor. From the experimental analysis, it is evident that the proposed work outperforms the analogous techniques in terms of accuracy, sensitivity and specificity.

Birdsongs recognition based on ensemble ELM with multi-strategy differential evolution

“…Venkatalakshmi et al extracted breast X-ray image feature set and combined ELM classifier to classify normal, malignant and benign breast cancer. The accuracy, sensitivity and specificity of the method are better than that of similar technology 9 . Kashif et al proposed an ELM-based consonant phoneme recognition model for the accent recognition of different pronunciations of English consonant phonemes by native Arabic speakers, the accuracy of the model reached 88% 10 .…”