Lung Anomaly Detection System (LADS) Using SVM based on Firefly Algorithm

Computed Tomography is one of the most popular technique which is used for lung anomaly detection. In this paper, an automatic lung anomaly detection system is developed by using SVM and Firefly optimization algorithm. For that features from the CT images are selected from various types of feature selection techniques such as Local Binary Pattern (LBP), Discrete Wavelet Transform (DWT) and Histogram of Oriented Gradients (HOG). Then the selected feature set are optimized by firefly optimization, to find out an optimal feature subset from candidate features. Then the feature set is filtered by a guided filter, which is mainly used in medical applications to improve the accuracy of our system. The filtered features are then classified by using a commonly used classifier SVM (Support Vector Machine) which classify the lung disease from the CT image. This paper further compare LADS with FIG method and LADS shows better recognition performance.