Comparative Analysis of SVM and CNN for Pneumonia Detection in Chest X-Ray

Ni Wayan Sumartini Saraswati (Institut Bisnis dan Teknologi Indonesia)
Dewa Ayu Putu Rasmika Dewi (Department of Infectious Diseases, School of Medicine, International University of Health and Welfare, Japan)
Poria Pirozmand (Faculty of Higher Education, Holmes Institute, Sydney, NSW 2000, Australia)

Lontar Komputer: Jurnal Ilmiah Teknologi Informasi
Vol 15 No 1 (2024): Vol. 15, No. 1 April 2024

Recognizing pneumonia sufferers can be done by analyzing chest X-ray images. Pneumonia sufferers experience pleural effusion, where fluid is between the lungs’ layers. It causes the lungs’ X-ray picture to be cloudy or hazy. It differs from the appearance of X-rays on normal lungs which are dark in color. These differences in X-Ray images can be classified automatically with the help of Artificial Intelligence This research used convolutional neural networks and support vector machine methods to recognize X-ray images of pneumonia. This research applied Principal Component Analysis and Wavelet Transformation support to both methods. This research aimed to evaluate the performance of each model combination. The PCA-SVM model gave the best performance, with an accuracy of 94.545% and an F1 score of 94.675%. The SVM model outperforms the CNN model in recognizing images; in this case, it could be due to the relatively small amount of training data.