Particle swarm optimization for segmentation of microcalcifications in mammograms

Breast cancer continues to be a significant public health problem in the world. Early detection is the key to improve the breast cancer prognosis. Microcalcification is one of the reliable methods for early detection of breast cancer. In this paper, detection of microcalcification is performed in two steps: enhancement and segmentation. The global appearance (brightness, contrast, etc.) of the breasts may differ, usually due to variations in the recording procedure. Initially the mammogram images are enhanced by median filter to remove the high frequency components (i.e. noise) from the image. The enhanced mammogram obtained from median filtering technique is evaluated by signal to noise ratio and compared with the existing techniques. The enhanced mammogram images are normalized to avoid differences in brightness between the mammograms caused by the recording procedure. The pectoral muscle region is removed from the breast region to increase the reliability of segmentation. Segmentation is performed using a novel method, Markov Random Field (MRF) hybrid with Particle Swarm Optimization (PSO) algorithm implemented to extract the suspicious region from background tissue. A recent population-based approach is used to segment the microcalcifications from the mammogram image.