Automated Detection of Viral Lung Infections from CT Scans Using Deep Learning

This study presents a deep convolutional neural network (CNN) for automated detection of viral lung infections from CT scans. The model was trained on a curated dataset of 5,216 scans and achieved 98% accuracy, significantly outperforming traditional methods like SVM and Random Forest. The lightweight architecture ensures high processing speed (<2 seconds per scan) and deployability across web and mobile platforms, supporting rapid and reliable clinical diagnosis. Future work includes integration of Explainable AI (e.g., Grad-CAM) for clinical transparency and adaptability to emerging viral infections.