Related Experiment Video
Updated: Aug 30, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
851
Deep Convolutional Neural Network Mechanism Assessment of COVID-19 Severity
J Nirmaladevi1, M Vidhyalakshmi2, E Bijolin Edwin3
1Department of Information Science and Engineering, Bannari Amman Institute of Technology, Sathyamangalam, Tamil Nadu 638401, India.
Biomed Research International
|September 2, 2022
Summary
This study introduces a deep convolutional neural network (CNN) for assessing COVID-19 severity using chest X-rays. The model accurately classifies patients into four risk categories, aiding in pandemic management.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Infectious Diseases
Background:
- COVID-19 diagnosis faces challenges with current testing methods.
- Patient overload in radiology departments impacts care quality.
- Accurate patient risk stratification is crucial for managing COVID-19.
Purpose of the Study:
- To develop a novel deep convolutional neural network (CNN) model for assessing COVID-19 severity.
- To classify COVID-19 patients into distinct risk categories using chest X-ray images.
Main Methods:
- An unsupervised deep convolutional neural network (DCNN) model was developed.
- Chest X-ray images were utilized as input data.
- Hyperparameters were dynamically adjusted using variable selection optimization.
Main Results:
- The DCNN model achieved 96% accuracy in classifying COVID-19 severity.
- Patients were categorized into four risk levels: low, medium, serious, and critical.
- Empirical validation was performed on a large dataset of chest X-ray scans.
Conclusions:
- This research presents the first multi-phase COVID-19 severity assessment using a large X-ray dataset and a DCNN.
- The DCNN model offers a promising tool for objective COVID-19 risk stratification.
- The findings support the use of AI in improving diagnostic accuracy and patient management during pandemics.
