Related Experiment Video
Updated: Aug 2, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.2K
Predicting the Severity of COVID-19 from Lung CT Images Using Novel Deep Learning.
Ahmad Imwafak Alaiad1, Esraa Ahmad Mugdadi1, Ismail Ibrahim Hmeidi1
1Computer Information System, Jordan University of Science and Technology, Irbid, Jordan.
Journal of Medical and Biological Engineering
|April 20, 2023
Summary
A deep learning model accurately predicts COVID-19 severity from lung CT scans, achieving 99.5% accuracy. This advancement aids in diagnosing and managing coronavirus disease 2019 (COVID-19) patients.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Infectious Diseases
Background:
- Coronavirus 2019 (COVID-19) has caused significant global health and economic disruption.
- While qRT-PCR detects infection, it is insufficient for assessing COVID-19 severity and lung involvement.
- Lung CT scans offer a visual assessment of disease impact.
Purpose of the Study:
- To develop a deep learning model for predicting COVID-19 severity using lung CT images.
- To enhance diagnostic capabilities for coronavirus disease 2019 (COVID-19).
Main Methods:
- A dataset of 2205 lung CT images from 875 COVID-19 patients was collected.
- Images were classified by a radiologist into normal, mild, moderate, and severe categories.
- Multiple deep learning algorithms were evaluated for severity prediction.
Main Results:
- The Resnet101 deep learning algorithm demonstrated superior performance.
- The model achieved an accuracy of 99.5% in predicting COVID-19 severity.
- A low data loss rate of 0.03% was recorded.
Conclusions:
- The developed deep learning model shows high efficacy in assessing COVID-19 severity from CT scans.
- This tool can support the diagnosis and treatment of COVID-19 patients.
- Improved patient outcomes are anticipated through the application of this predictive model.
