CONVOLUTIONAL NEURAL NETWORK MODEL APPLIED TO PNEUMONIA DETECTION

Abstract

Deep learning has become increasingly applicable across various domains, with the recent surge in available data, especially
within the medical field. Its primary role is to enhance decision-making by making it more efficient, accurate, and reliable. Both deep learning and machine learning are growing applications in medicine. This is especially evident in medical areas involving different biomedical imaging types and relying heavily on collecting and analyzing vast quantities of digital images. This paper explores the utilization of advanced deep learning models, specifically convolutional neural networks, to analyze chest X-ray images and aid in accurate diagnosis. The convolutional neural network model is designed to tackle the classification challenge of identifying whether
chest X-ray images indicate the presence of pneumonia, utilizing a dataset that includes both normal chest X-rays and those exhibiting signs of viral pneumonia. The study experimentally applies and assesses the accuracy of the VGG16 and ResNet50 models.