Detection of Pneumonia Disease Using Deep Learning in Chest X-Ray Dataset

Abstract

Early diagnosis is very important in pneumonia, which is frequently encountered in the world and sometimes leads to fatal results. Chest X-rays are often preferred in diagnosing pneumonia. Deep learning and machine learning algorithms, which are special usage models of artificial intelligence, have recently been frequently preferred in the detection of pneumonia. In this study, convolutional neural network (CNN), which is preferred in the field of image processing, was chosen for pneumonia diagnosis. The chest X-ray dataset shared with the public by the NIH Clinical Center was used. In the experimental study, the feature selection process was performed with CNN, while linear regression (LR), support vector machine (SVM), and decision tree (DT) were used in the classification phase. The values obtained by these classifiers were compared. As a result of the study, the highest value belongs to the SVM classifier with 96.4%. It was observed in this study that the CNN algorithm achieved high performance in the feature extraction process from image expressions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.