Lung CT image–based diagnosis of COVID-19 infection is an important research topic in medical image analysis. However, there has been little research on diagnostic support systems that combine deep learning and parallel computing. In this paper, a diagnostic support system based on a deep learning framework is proposed. Specifically, the YOLOv5 model is trained on a public COVID-19 lung CT dataset, and GPU acceleration using CUDA and cuDNN is applied to significantly reduce training time. Through the experimental results, the proposed system shows good performance in both diagnostic accuracy and computational efficiency.
