Deep Bidirectional Classification Model for COVID-19 Disease Infected Patients
Summary
A novel deep learning model, MADE-DBM, effectively classifies COVID-19 from chest CT scans. This approach shows superior performance compared to existing methods, aiding in early diagnosis of the novel coronavirus infection.
Area of Science:
- Medical Imaging
- Artificial Intelligence
- Infectious Diseases
Background:
- The COVID-19 pandemic, caused by a novel coronavirus, emerged in late 2019.
- Chest CT scans are crucial for early COVID-19 detection, complementing RT-PCR tests.
- Accurate classification of COVID-19 from CT images presents a significant challenge.
Purpose of the Study:
- To propose a deep learning model for automated COVID-19 classification from chest CT images.
- To enhance classification accuracy and efficiency for early detection of novel coronavirus infection.
- To introduce a hybrid approach combining deep bidirectional long short-term memory networks with a mixture density network (DBM).
Main Methods:
- Development of a deep bidirectional long short-term memory network with mixture density network (DBM) model.
- Utilization of a Memetic Adaptive Differential Evolution (MADE) algorithm for optimizing DBM hyperparameters.
- Experimental validation using benchmark chest CT image datasets.
Main Results:
- The proposed MADE-DBM model demonstrated superior performance in COVID-19 classification.
- Comparative analysis confirmed the effectiveness of MADE-DBM over existing approaches.
- The model achieved high accuracy across various performance metrics.
Conclusions:
- The MADE-DBM model offers a robust and efficient solution for real-time COVID-19 classification.
- This deep learning approach can significantly aid radiologists and clinicians in diagnosing COVID-19.
- The study highlights the potential of AI in managing infectious disease outbreaks through medical imaging analysis.
