An In-Depth Analysis of Pneumonia Detection Utilizing Information GAN (InfoGAN) and Convolutional Neural Networks (CNN): A Comprehensive Deep Learning Framework

Abstract

Automation of pneumonia detection is essential in the medical industry. This concept offers a promising approach to the modern era’s pneumonia detection. Compared to CT scans, pneumonia is more difficult to identify in X-ray pictures. However, compared to X-rays, CT scans require more radiation throughout the operation. Enhancing the odds of survival is largely dependent on early identification of lung infection. In this research, Keras and Tensorflow are used to identify pneumonia and extract lung features. Following feature extraction, the model is pre-processed to obtain an algorithm overview. Following pre-processing, the model feeds the Info GAN algorithm with the characteristics that were collected from the data. The discriminator verifies the erroneous data generated by the algorithm. The model is constructed if every sample and fake case is tested. The model will be coupled with a GUI after it has been trained. An Input can be given from the GUI to the model so that it can check for pneumonia.