Lung Nodule Detection and Classification based on Feature Merging and Genetic Algorithm

Abstract

The purpose of this study is to increase the accuracy of the early detection of lung nodes through the design of Computer-Aided Detection (CAD). Comparative studies of the effectiveness of the most commonly used methods for extracting and classifying features were conducted, and methods that give the highest accuracy and the least false-positive results were identified. To develop an accurate automatic detection system for early detection of lung nodules, we investigated directly candidates for lung nodules without removing blood vessels. The feature is extracted by using feature extraction methods. To utilize the extracted features, the extracted features were connected using the feature merge technique, and the features were selected by the new hybrid feature vector. The genetic algorithm (GA) search based on the classification accuracy rate of the used classifier was also applied to the hybrid feature vector. To archive, high classification accuracy was selected three classifiers and performed a comparison of their performance.