Related Experiment Video
Updated: Oct 20, 2025
04:17
DNA Virus Detection System Based on RPA-CRISPR/Cas12a-SPM and Deep Learning
Published on: May 10, 2024
1.0K
CovH2SD: A COVID-19 detection approach based on Harris Hawks Optimization and stacked deep learning
Hossam Magdy Balaha1, Eman M El-Gendy1, Mahmoud M Saafan1
1Computers Engineering and Systems Department, Faculty of Engineering, Mansoura University, Egypt.
Expert Systems with Applications
|September 13, 2021
Summary
A new hybrid deep learning approach, CovH2SD, accurately detects COVID-19 from chest CT scans. This method optimizes hyperparameters using Harris Hawks Optimization, outperforming existing studies in speed and accuracy for rapid diagnosis.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Computational Biology
Background:
- Coronavirus disease (COVID-19) rapidly spread globally, causing significant mortality.
- Accurate and timely diagnosis of COVID-19 is critical for containment.
- Reverse Transcription Polymerase Chain Reaction (RT-PCR) tests have limitations in speed, accuracy, and availability.
Purpose of the Study:
- To propose a novel hybrid learning and optimization approach for COVID-19 detection using chest Computed Tomography (CT) images.
- To enhance the accuracy and efficiency of COVID-19 diagnosis through advanced computational methods.
Main Methods:
- Developed CovH2SD, a hybrid approach combining deep learning feature extraction with Harris Hawks Optimization (HHO) for hyperparameter tuning.
- Employed transfer learning with nine pre-trained convolutional neural networks (ResNet50, ResNet101, VGG16, VGG19, Xception, MobileNetV1, MobileNetV2, DenseNet121, DenseNet169).
- Utilized Fast Classification Stage (FCS) and Compact Stacking Stage (CSS) to integrate optimal models.
Main Results:
- Achieved a top Weighted Sum Method (WSM) value of 99.31% and accuracy of 99.33%.
- Six experiments demonstrated WSM values exceeding 96.5%.
- The proposed method outperformed eleven state-of-the-art studies in COVID-19 detection accuracy.
Conclusions:
- CovH2SD offers a highly accurate and efficient method for COVID-19 detection from CT images.
- The hybrid deep learning and optimization strategy shows significant promise for improving diagnostic capabilities.
- This approach can aid in faster and more reliable identification of COVID-19 patients.
