Analysis of Chest X-Ray Images and Detection of Pneumonia Using Deep Learning

EasyChair Preprint 8316

Abstract

On account of pneumonia, the lungs are impacted by a bacterial contamination. An early examination is critical to a successful cure. In most cases, an expert radiologist can determine the dilemma with a chest X-ray. The visualization can be self-endorsing for a spread of reasons, for example, a strange look on the chest X-beam photographs or perhaps muddled with different sicknesses. As a result of this, practitioners will need to be guided by computer-aided diagnosing tools. The models Convolutional Neural Networks, VGG16 and Inception -V3 wear utilized in this research study were deep rooted for diagnosing pneumonia. Investigating results showed that the Vgg16 and Inception-V3 models were having an accuracy of 0.80%, and 0.76% individually. However, when compared to the Inception-V3 model, the Vgg16 model proven more effective in detecting pneumonia patients. This study demonstrates that each model has its own specialty and capabilities for the same dataset.

Keyphrases: Chest X-Ray., Convolutional Neural Networks, In-ception V3, Pneumonia detection, Transfer Learning, VGG-16, deep learning

@booklet{EasyChair:8316,
  author    = {Varsha Dange and Varun Patil and Shreya Pawar and Yukta Pedhavi and Parikshit Pohane and Pranita Maske},
  title     = {Analysis of Chest X-Ray Images and Detection of Pneumonia Using Deep Learning},
  howpublished = {EasyChair Preprint 8316},
  year      = {EasyChair, 2022}}