Deepal, D. A. A., & Fernando, T. G. I. (2021). Convolutional Neural Network Approach for the Detection of Lung Cancers in Chest X-Ray Images. In U. Kose & J. Alzubi (Eds.), Deep Learning for Cancer Diagnosis (pp. 203–226), Studies in Computational Intelligence (SCI) Book Series, Vol. 908. Springer. https://doi.org/10.1007/978-981-15-6321-8_12
Abstract:
Chest X-rays are considered to be the most widely used technique within the health industry for the detection of lung cancers. Nevertheless, it is very difficult to identify lung nodules using raw chest X-ray images and analysis of such medical images has become a very complicated and tedious task. This study mainly concerned on convolutional neural network approach to identify whether a suspicious area is a nodule or a non-nodule. The JSRT digital images of the chest X-ray database developed by the Japanese Society of Radiological Technology (JSRT) is used to train and test these models. Further, support vector machines and multilayer perceptrons are used for comparison with convolutional neural network model. “Pylearn2” research library is used to build the convolutional neural network model and multilayer perceptron model. “scikit-learn” Python library is used to build the support vector machine models. “MATLAB” is used to extract nodule and non-nodule locations from the original images and other image processing parts. Under support vector machine models, three functions (linear, polynomial and radial) are used and the linear function showed the highest accuracy rate (92%). Comparing of these three approaches, the convolutional neural network approach showed the highest accuracy rate (94%).
Keywords:
Convolutional neural network (CNN), Graphics processing unit (GPU), Gray level co-occurrence matrix (GLCM), Japanese society of radiological technology (JSRT), Multilayer perceptron (MLP), Support vector machine (SVM)