Abstract <p>Deep neural networks have enabled significant breakthroughs in medical image analysis. However, due to their high data requirement, small datasets in medical imaging tasks may prevent them from reaching their full potential. Synthetic data generation is a promising alternative to augment training datasets and enable medical image research on a larger scale. Recently, diffusion models have attracted the attention of the computer vision community because of their ability to generate photorealistic synthetic images. In this paper, we explore the possibilities of using diffusion models and develop computer algorithms for creating high-resolution synthetic images.</p>