Transfer learning based novel ensemble classifier for COVID-19 detection from chest CT-scans

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Infectious Diseases

Background:

• Coronavirus Disease 2019 (COVID-19) poses a significant global health challenge, necessitating rapid and accurate diagnostic methods.
• Automating COVID-19 detection from medical images, particularly Lung Computed Tomography (CT) scans, is crucial for pandemic control.
• Existing deep learning models for COVID-19 detection often rely on single predictions, potentially limiting accuracy.

Purpose of the Study:

• To develop and evaluate a novel ensemble deep learning approach for enhanced COVID-19 detection from Lung CT scan images.
• To aggregate the predictive power of multiple pre-trained deep neural network architectures for improved diagnostic performance.
• To establish a new state-of-the-art in automated COVID-19 detection using medical imaging.

Main Methods:

• Utilized various pre-trained deep neural network models including VGG16, VGG19, InceptionV3, ResNet50, ResNet50V2, InceptionResNetV2, Xception, and MobileNet.
• Fine-tuned these models using a dataset of Lung CT scan images.
• Developed an ensemble classifier by aggregating predictions from the fine-tuned models to make the final diagnosis.

Main Results:

• The proposed ensemble approach demonstrated superior performance compared to existing methods.
• The ensemble model achieved state-of-the-art results in detecting COVID-19 infection from Lung CT scan images.
• Aggregating multiple deep learning models significantly improved the accuracy of COVID-19 diagnosis.

Conclusions:

• Ensemble deep learning methods offer a robust and accurate solution for COVID-19 detection from Lung CT scans.
• The proposed novel ensemble approach sets a new benchmark for automated COVID-19 diagnosis in medical imaging.
• This strategy effectively overcomes the limitations of single-model predictions in identifying COVID-19 infection.