A Weighted Particle Swarm Optimization (WPSO) technique for accurate breast cancer detection

Abstract

Microcalcifications are one of the key symptoms facilitating early detection of breast cancer. In this paper,
textural features are extracted from the segmented mammogram image to classify the microcalcifications into benign,
malignant or normal. The tumor part of the breast region is extracted by the reduction algorithm. Initially the reduced
features are normalized between zero and one. The normalized feature values are given as input to a three-layer BPN to
classify the microcalcifications into benign, malignant or normal. The network is trained to produce the output value for
normal images. In this research proposed an image extraction method with reduced computation time was proposed. The
computational time of proposed approach is obtained by use of Particle Swarm Optimization (PSO) algorithm. The
membership function of this conventional algorithm is sensitive to the outline and does not integrate the spatial
information in the image. To resolve existing disadvantages in classification algorithms like sensitivity, noise and
in-homogeneties are overcome in proposed approach. Existing drawbacks are overcome in our algorithm takes also into
consideration the spatial neighborhood information. To resolve existing limitation in this research proposed a weighted
particle swarm optimization technique for cancer extraction. Based on the weighted function in optimization algorithm
images are clustered and extracted. Simulation results of the proposed approach are comparatively examined with
existing approaches naive bayes, Adaboost and SVM. Comparison of results demonstrated that proposed approach
exhibits effective performance than existing classification algorithm.