Pneumonia Detection and Classification using Hybrid Convolution Neural Network and Machine Learning Classifiers

Abstract

Viruses, bacteria, and fungus are the main causes of pneumonia, an infectious disease that affects the lungs in people. It is mostly seen individuals older than sixty years of age and youngsters younger than five. Pneumonia causes fever, shortness of breath, nausea, diarrhea, etc. So, in this project Hybrid Convolution Neural Network (CNN) is used with Machine Learning Classifiers to detect Pneumonia disease from the Chest X-rays, which are used in the real world by medical professionals to find pneumonia. Computer-Aided Diagnosis (CAD) systems may be improved with the application of deep learning and machine learning based technologies, which can be of assistance to radiologists and doctors when making medical decisions. CNN have shown a significant amount of promise in terms of image classification and segmentation, and they are often used for the development of DL-based CAD systems. The findings indicate that the suggested ensemble classifier, which combines Radial Basis Function (RBF), classifiers using support vector machines (SVM) and logistic regression (LR), performs best, with an accuracy rate of 97%. Ultimately, a web-based CAD system is made using this concept that will greatly enhance radiologists' capacity to recognize pneumonia.