Measurement-conditioned Denoising Diffusion Probabilistic Model for Under-sampled Medical Image Reconstruction
This work addresses the challenge of accelerating MRI reconstruction for medical imaging applications, representing an incremental improvement by adapting diffusion models to a specific domain.
The paper tackles the problem of under-sampled medical image reconstruction by proposing a measurement-conditioned denoising diffusion probabilistic model (MC-DDPM) that operates in the measurement domain and is conditioned on under-sampling masks, achieving excellent performance that outperforms full supervision baseline and state-of-the-art score-based methods in MRI reconstruction.
We propose a novel and unified method, measurement-conditioned denoising diffusion probabilistic model (MC-DDPM), for under-sampled medical image reconstruction based on DDPM. Different from previous works, MC-DDPM is defined in measurement domain (e.g. k-space in MRI reconstruction) and conditioned on under-sampling mask. We apply this method to accelerate MRI reconstruction and the experimental results show excellent performance, outperforming full supervision baseline and the state-of-the-art score-based reconstruction method. Due to its generative nature, MC-DDPM can also quantify the uncertainty of reconstruction. Our code is available on github.