Detection of Microcalcifications Using Pixcals Refined Bandwidth Algorithm in Digitized Mammograms
Abhisek Santra, W. Jai Singh, S. Arul · International journal of tomography and simulation · 2011
Clustered microcalcifications are one of the earliest indicators of breast cancer and can be detected only by mammography. Microcalcifications appear in a mammogram as fine, granular clusters, which are often difficult to identify in a raw mammogram. For early breast cancers, screening studies suggest that 70 to 90 percent were detected based on microcalcifications alone. The clinical procedure is difficult, time consuming and demands great concentration during screening. Hence an attempt has been made to identify the suspicious region by a Computer Aided System. A Pixcals Refined Bandwidth algorithm is being developed which moves over pixels in order to extract microcalcification region. The new algorithm is embedded in a system which will enable to classify an image as normal or abnormal. Also for an abnormal image it indicates and earmarks the suspected area which contains microcalcifications. Different kinds of images have been considered and tested using the proposed algorithm and results are given in tables. The efficiency is higher which is measured through Receiver Operating Characteristic (ROC) curve.