Towards better veracity for breast cancer detection using Gabor analysis and statistical learning

Breast Cancer is by far the most prevalent cancer diagnosed in women worldwide. Early diagnosis and detection is now possible through modern technology like mammography. In this paper, we present a method to augment the detection process by efficiently recognizing the carcinogenic tissue or cells. To address this issue, we propose an algorithm using Discrete Gabor Wavelet Transforms based on Hidden Markov Model for classification. We test our proposed method on the Mini Mammographie Image Analysis Society (MIAS) database. The proposed method yields about 90% recognition accuracy. This increase in accuracy is due to the statistical classification of benign and malignant cells.