Abstract—Pneumonia commonly results from a lung ailment that leads to irritation and harm to the lungs. A chest X-ray is one of the most effective imaging techniques for detecting pneumonia, but diagnosing and treating it can be difficult due to its similarity to other lung conditions. To improve the accuracy of classifying X-ray images, we suggest using an ensemble model in our research that combines deep Convolutional Neural Network (CNN) architectures. The suggested approach classifies the input image as having pneumonia or not by extracting data features using an ensemble of three CNN models. The comparison involves using a single CNN model and a combination of CNN models to evaluate the ensemble architecture. This work evaluates the InceptionResNetV2, DenseNet201, and VGG16 ensemble. The suggested ensemble algorithm provides comparatively positive classification results with an accuracy of almost 95%, outperforming previous ensemble models and improving the average F1-Score by 3% compared to the single model approach.
 
Keywords—pneumonia, chest X-ray, ensemble learning, deep learning, convolutional neural network


Copyright © 2024 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.