Related Experiment Video
Updated: Aug 19, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
842
Cloud-based COVID-19 disease prediction system from X-Ray images using convolutional neural network on smartphone
Madhusudan G Lanjewar1, Arman Yusuf Shaikh1, Jivan Parab1
1School of Physical and Applied Sciences, Goa University, Taleigao Plateau, Goa, 403206 India.
Multimedia Tools and Applications
|December 5, 2022
Summary
A novel Convolutional Neural Network (CNN) system effectively predicts COVID-19 from Chest X-Ray (CX-Ray) images. This automated approach offers a faster, more accurate alternative to RT-PCR, reducing healthcare system strain.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Infectious Disease Diagnostics
Background:
- COVID-19 diagnosis relies heavily on RT-PCR, which has limitations including low sensitivity and time constraints.
- Chest X-Ray (CX-Ray) imaging reveals significant manifestations of coronavirus infections, presenting an alternative diagnostic avenue.
- Automated COVID-19 detection systems can alleviate pressure on healthcare infrastructure.
Purpose of the Study:
- To develop and evaluate a real-time Convolutional Neural Network (CNN) based system for COVID-19 detection using CX-Ray images.
- To compare the performance of the developed CNN model against deep Convolutional Neural Networks (DCNNs) like ResNet50, VGG19, InceptionV3, and Xception.
- To deploy the efficient CNN model on a cloud platform for accessible and scalable use.
Main Methods:
- A Convolutional Neural Network (CNN) model was implemented for real-time COVID-19 prediction from CX-Ray images.
- The CNN model's performance was assessed using metrics including accuracy, precision, recall, F1 score, and Area Under the Curve (AUC).
- Comparative analysis was conducted against several DCNN architectures (ResNet50, VGG19, InceptionV3, Xception) using the same CX-Ray dataset.
Main Results:
- The implemented CNN model achieved high performance, with 99.94% training accuracy and 98.81% validation accuracy.
- Evaluation metrics demonstrated excellent results: 99% precision, 98% recall, 99% F1 score, 100% training AUC, and 98.3% validation AUC.
- The developed CNN model outperformed the tested DCNN architectures in COVID-19 prediction accuracy.
Conclusions:
- The developed real-time CNN system provides a highly accurate and efficient method for COVID-19 detection from CX-Ray images.
- This AI-driven approach shows superior performance compared to established DCNN models, offering a promising alternative diagnostic tool.
- The successful deployment of the CNN model on a cloud platform (PaaS) enables wider accessibility and application in healthcare settings.
