Binary Classification of Pneumonia Disease Using SVM-GAN Augmentation

Pneumonia is a respiratory disease that causes inflammation of the lungs due to the accumulation of fluid in the air sacs of the alveoli. This disease can affect one or both lungs. According to a study by the World Health Organization, all children, newborns and people over the age of 65 suffer from this disease. Deep learning algorithms reduce the workload of medical professionals by automating the necessary procedures, since physical examinations by radiologists do not always provide accurate results in the process of disease detection and are also somewhat time-consuming. This article discusses the classifications of pneumonia, and non-pneumonia. The most frequent diagnostic method for these disorders is chest X-rays, which are utilized as scanned pictures. The SVM and GAN technique was used to classify the disease. The photos that were gathered were acquired from online retailers, and the present data volume was raised using the GAN algorithm, which helped to train and test the model more successfully in the future using SVM. Combined, a better accuracy of 98.5% than that of previous methods was attained. As a result, the suggested approach makes it possible to identify pneumonia more rapidly and to classify the condition even from a limited quantity of data. It can be a useful tool for radiologists and other medical professionals working in this area.