Pneumonia Disease Classification Using Xception CNN Model

Abstract

Medical disease known as pneumothorax is characterized by air being present in the pleural cavity, which is the area of the chest that surrounds the lungs. This illness can cause the lung to partially or completely collapse, upsetting the normal balance between the pressures in the lung and the chest wall. Traumatic pneumothorax, which permits air to enter the pleural space, can be brought on by trauma such as rib fractures or penetrating chest injuries. Conversely, spontaneous pneumothorax. It can happen for no obvious reason and is frequently associated with lung disorders such as tiny air sac rupture in the lung. This paper uses the Xception Convolutional Neural Network (CNN) model to classify pneumonia infections. Pneumonia is a dangerous respiratory condition that requires prompt diagnosis to ensure appropriate treatment. The Xception architecture, which is an expansion of the Inception model and is well-known for its effectiveness in capturing complex visual information using depth-wise separable convolutions, is completely utilized in this study. The major objective is to develop a robust machine learning model that, using images from chest X- rays, can reliably classify individuals as having pneumonia or not. It includes putting together and preparing a labeled dataset, adjusting the Xception model, then training and verifying it thoroughly in order to maximize accuracy. Using deep learning methods, specifically the Xception CNN model, has the potential to improve pneumonia diagnosis and aid in more accurate and efficient medical decision-making

Ähnliche Arbeiten