Detection of lung cancer using radiograph images enhancement and radial basis function classifier

A lung cancer can be detected using the chest radiograph. In this study, images enhancement of high frequency emphasis filter and histogram equalization were applied to enhance the chest radiograph. This stage aimed to obtain higher performance of a lung cancer detection process. The radial basis function neural network (RBFNN) classifier was implemented to detect whether the lung has normal or cancer condition. The detection process of RBFNN was performed by evaluating the features extracted from the chest radiograph using the Gray Level Co-occurrence Matrix method. We involved five features, namely energy contrast, correlation, inverse difference moment, and entropy. The result demonstrated the effectiveness of image enhancement of high frequency emphasis filter and histogram equalization for increasing the accuracy of RBFNN classifier to detect a lung cancer using the chest radiograph.