The paper "Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image Generation" introduces Fast-DDPM, a method that significantly reduces the computational cost of training and sampling diffusion models (DDPMs) in medical imaging. Traditional DDPMs, which require a large number of time steps (e.g., 1,000) and can be computationally expensive, are underutilized in medical imaging due to their high computational demands. Fast-DDPM addresses this issue by training and sampling using only 10 time steps, optimizing time-step utilization. The key innovation is the introduction of two efficient noise schedulers: one with uniform sampling and another with non-uniform sampling. These schedulers are designed to align training and sampling procedures, ensuring that the denoiser is trained on the most relevant time steps. The method is evaluated on three medical image-to-image tasks: multi-image super-resolution, image denoising, and image-to-image translation. Fast-DDPM outperforms existing methods in all tasks, reducing training time by 0.2× and sampling time by 0.01× compared to DDPM. The code for Fast-DDPM is publicly available, making it accessible for further research and practical applications in medical imaging.The paper "Fast-DDPM: Fast Denoising Diffusion Probabilistic Models for Medical Image-to-Image Generation" introduces Fast-DDPM, a method that significantly reduces the computational cost of training and sampling diffusion models (DDPMs) in medical imaging. Traditional DDPMs, which require a large number of time steps (e.g., 1,000) and can be computationally expensive, are underutilized in medical imaging due to their high computational demands. Fast-DDPM addresses this issue by training and sampling using only 10 time steps, optimizing time-step utilization. The key innovation is the introduction of two efficient noise schedulers: one with uniform sampling and another with non-uniform sampling. These schedulers are designed to align training and sampling procedures, ensuring that the denoiser is trained on the most relevant time steps. The method is evaluated on three medical image-to-image tasks: multi-image super-resolution, image denoising, and image-to-image translation. Fast-DDPM outperforms existing methods in all tasks, reducing training time by 0.2× and sampling time by 0.01× compared to DDPM. The code for Fast-DDPM is publicly available, making it accessible for further research and practical applications in medical imaging.