R&D(2025–2026) Research on the Development of a Deep Learning–Based Image Translation Model for Converting Non-Contrast CT Images to Contrast-Enhanced CT Images
This study aims to develop a deep learning–based image translation model that converts non-contrast CT images into contrast-enhanced CT images. The model is designed to learn the relationship between non-contrast CT and contrast-enhanced CT images and generate synthetic images reflecting contrast enhancement effects. Through this approach, medical images that clearly represent vascular and lesion information can be produced without the use of contrast agents. Training strategies are also applied to maintain structural consistency and detailed tissue representation in the generated images. This technology is expected to provide alternative imaging information for patients who cannot receive contrast agents and contribute to reducing examination time and medical costs.
Project Period
2025~2026
This study aims to develop a deep learning–based image translation model that converts non-contrast CT images into contrast-enhanced CT images. The model is designed to learn the relationship between non-contrast CT and contrast-enhanced CT images and generate synthetic images reflecting contrast enhancement effects. Through this approach, medical images that clearly represent vascular and lesion information can be produced without the use of contrast agents. Training strategies are also applied to maintain structural consistency and detailed tissue representation in the generated images. This technology is expected to provide alternative imaging information for patients who cannot receive contrast agents and contribute to reducing examination time and medical costs.