The effectiveness of combining the likelihood maps of different filters in improving detection of calcification objects

Breast cancer is the most prevalent form of cancer diagnosed in women. Mammograms offer the best option in detecting the disease early, which allows early treatment and by implication, a favorable prognosis. This study looks to combine the Wavelet, Median, Gaussian and a Finite Impulse Response filters for the task of detecting Malignant and Benign calcifications, which are among the primary indicators of breast cancer in digital Mammograms. These filters individually detect calcifications to varying degrees of success, but also create artifacts especially along the boundaries of curvilinear structures. They are combined in a way that improves overall detection, while diminishing their individual side effects. An Entropy-based thresholding technique is finally used to segment the calcifications from the background. Experimental results show that the proposed model achieves a 100% detection rate, which shows the effectiveness of combining the likelihood maps from various filters in detecting calcification objects.