Related Experiment Video
Updated: Jun 16, 2025
08:21
Cell-Free Dot Blot as a Practical and Adaptable Immunoassay Platform for the Detection of Antibody Response in Human and Animal Sera
Published on: May 23, 2025
159
COVID-19 IgG antibodies detection based on CNN-BiLSTM algorithm combined with fiber-optic dataset
Mohammed Jawad Ahmed Alathari1, Yousif Al Mashhadany2, Ahmad Ashrif A Bakar1
1UKM - Department of Electrical, Electronic and Systems Engineering, Faculty of Engineering and Built Environment, Universiti Kebangsaan Malaysia, UKM, Bangi 43600, Malaysia.
Journal of Virological Methods
|August 18, 2024
Summary
This study introduces a hybrid deep learning model combining CNN and Bi-LSTM for COVID-19 detection using IgG antibody data. The CNN-BiLSTM model achieved high accuracy, showing promise for automated COVID-19 screening.
Area of Science:
- Biomedical Engineering
- Artificial Intelligence in Healthcare
- Infectious Disease Diagnostics
Background:
- The COVID-19 pandemic necessitates rapid and accurate diagnostic tools.
- Automated screening methods are crucial for efficient disease detection.
- Current diagnostic approaches require improvement in speed and accuracy.
Purpose of the Study:
- To develop and evaluate a hybrid deep learning model for COVID-19 detection.
- To assess the performance of Convolutional Neural Network (CNN) and Bi-directional Long Short-Term Memory (Bi-LSTM) networks.
- To utilize fiber optic data with SARS-CoV-2 Immunoglobulin G (IgG) antibodies for enhanced diagnostics.
Main Methods:
- A hybrid CNN-BiLSTM model was developed for COVID-19 classification.
- Four deep learning algorithms (CNN, CNN-RNN, BiLSTM, CNN-BiLSTM) were evaluated.
- Fiber optic data and SARS-CoV-2 IgG antibody information were integrated into the model.
- A comprehensive data preprocessing pipeline was implemented.
Main Results:
- The CNN-BiLSTM model demonstrated superior performance among the evaluated algorithms.
- Achieved accuracy of 89%, recall of 88%, precision of 90%, and F1-score of 89%.
- Specificity reached 90%, G-mean was 89%, and ROC was 96% on training datasets.
Conclusions:
- The proposed CNN-BiLSTM model shows significant potential for accurate COVID-19 classification.
- The integration of IgG antibodies enhances diagnostic specificity and accuracy.
- This hybrid model could serve as a valuable automated screening tool for healthcare professionals.
