Neighborhood Rough Neural Network Approach for COVID-19 Image Classification

Area of Science:

• Medical Imaging
• Artificial Intelligence
• Computational Biology

Background:

• The rapid global spread of COVID-19 necessitates efficient diagnostic tools.
• Computed Tomography (CT) scans offer a viable method for rapid COVID-19 screening.
• Accurate classification of COVID-19 from medical images is crucial for patient management.

Purpose of the Study:

• To propose a novel hybridized classification approach for COVID-19 detection.
• To evaluate the efficacy of the proposed method against existing benchmark algorithms.
• To enhance the accuracy and precision of COVID-19 image classification.

Main Methods:

• A novel hybrid classification model combining Neighborhood Rough Set Classification (NRSC) and Backpropagation Neural Network (BPN).
• Implementation of NRSC and BPN for classifying chest CT scan images as COVID-19 positive or negative.
• Comparative analysis against benchmark methods including Decision Tree, Random Forest, Naive Bayes, K-Nearest Neighbor, and Support Vector Machine.

Main Results:

• The proposed hybridized NRSC-BPN method demonstrated superior performance in classifying COVID-19 and NON-COVID images.
• Quantitative assessment using various classification accuracy measures confirmed the efficacy of the novel approach.
• The hybrid model achieved higher accuracy compared to individual methods and other benchmark algorithms.

Conclusions:

• The novel hybridized NRSC-BPN approach provides an effective and accurate method for COVID-19 image classification.
• This AI-driven diagnostic tool has the potential to aid clinicians in faster and more reliable COVID-19 detection.
• Further research can explore the integration of this method into clinical workflows for improved public health outcomes.