Efficient Detection of COVID-19 and Pneumonia from Chest X-ray Images Using Deep Learning Techniques
1st Assistant Prof. A. V. Kolaki
Department of Electronic and Communication
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal
avkolaki@klsvdit.edu.in
2nd Ms. Priyanka S Hiremath
Department of Electronic and Communication
KLS Vishwanathrao Deshpande
Institute of Technology
Haliyal
priyankahiremath11@gmail.com
3rd Ms. Priya Gingaigol
Department of Electronic and Communication
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal
priyaginagaigol@gmail.com
4th Ms. Sakshi S Hiremath
Department of Electronic and Communication
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal
hiremathsakshi30@gmail.com
5th Ms. Tulasa Shivanand Patil
Department of Electronic and Communication
KLS Vishwanathrao Deshpande Institute of Technology
Haliyal
patiltulasa11@gmail.com
Abstract – Respiratory diseases such as COVID-19 and pneumonia remain a significant global health challenge, particularly in developing regions where access to experienced radiologists is limited. Chest X-ray (CXR) imaging is one of the most widely used diagnostic tools, but manual interpretation is time-consuming and prone to inconsistencies. This paper presents an integrated offline deep-learning system for automatic classification of CXR images into three categories—COVID-19, Pneumonia, and Normal. The proposed framework combines a lightweight Convolutional Neural Network (CNN) for classification and a U-Net architecture for lung segmentation to enhance interpretability. A desktop-based Tkinter GUI is designed to display the prediction, confidence scores, lung mask overlay, and diagnostic suggestions, along with a text-to-speech feature. Experiments on a publicly available Mendeley dataset demonstrate that the model achieves an accuracy of approximately 90%. The system functions entirely offline, making it suitable for rural healthcare environments and emergency screening applications.
Keywords— COVID-19, Pneumonia, Chest X-ray, Deep Learning, CNN, U-Net, Image Segmentation, Healthcare AI, Computer-Aided Diagnosis, Tkinter GUI.
