AI – Based Pneumonia Detection Using CNN and Transfer Learning

Authors

  • Vinodhini B Assistant professor, Dept. of AI&DS, Jai Shriram Engineering College, Tirupur, Tamilnadu, India Author
  • Akshay R UG Scholar, Dept. of AI&DS, Jai Shriram Engineering College, Tirupur, Tamilnadu, India Author
  • Gowtham B UG Scholar, Dept. of AI&DS, Jai Shriram Engineering College, Tirupur, Tamilnadu, India Author
  • Haripraneshwaran S UG Scholar, Dept. of AI&DS, Jai Shriram Engineering College, Tirupur, Tamilnadu, India Author
  • Jayavarman R UG Scholar, Dept. of AI&DS, Jai Shriram Engineering College, Tirupur, Tamilnadu, India Author

DOI:

https://doi.org/10.47392/IRJAEH.2026.0282

Keywords:

Pneumonia Detection, Chest X-ray Imaging, Deep Learning, Transfer Learning, Model Ensembling

Abstract

Pneumonia still remains one of the primary causes of health issues and fatalities worldwide since it requires immediate treatment in healthcare institutions that do not always have powerful diagnostic instruments. The process of reading the chest X-ray images is not only difficult, but also time-consuming, and results offer varied outcomes as observed by different observers. This paper will discuss a deep learning model that will only consume little computing power to automatically discover pneumonia using convolutional neural networks, transfer learning, and model ensemble methods. To make new models with better generalization and overfitting resistance the researchers used several lightweight pre-trained models including AlexNet, ResNet-18, DenseNet-201 and SqueezeNet. The experiment results indicate good results in terms of accuracy, sensitivity and F1-score and less processing need. These were improvements authenticated on openly available chest X-rays.

Downloads

Download data is not yet available.

Downloads

Published

2026-04-27

How to Cite

AI – Based Pneumonia Detection Using CNN and Transfer Learning. (2026). International Research Journal on Advanced Engineering Hub (IRJAEH), 4(04), 2121-2127. https://doi.org/10.47392/IRJAEH.2026.0282