DBSR: Quadratic Conditional Diffusion Model for Blind Cardiac MRI Super-Resolution

Defu Qiu,Yuhu Cheng,Kelvin K.L. Wong,Wenjun Zhang,Zhang Yi,Xuesong Wang

DOI: https://doi.org/10.1109/tmm.2024.3453059

IF: 7.3

2024-01-01

IEEE Transactions on Multimedia

Abstract:Cardiac magnetic resonance imaging (CMRI) can help experts quickly diagnose cardiovascular diseases. Due to the patient's breathing and slight movement during the magnetic resonance imaging scan, the obtained CMRI may be severely blurred, affecting the accuracy of clinical diagnosis. To address this issue, we propose the quadratic conditional diffusion model for blind CMRI super-resolution (DBSR). Specifically, we propose a conditional blur kernel noise predictor, which predicts the blur kernel from low-resolution images by the diffusion model, transforming the unknown blur kernel in low-resolution CMRI into a known one. Meanwhile, we design a novel conditional CMRI noise predictor, which uses the predicted blur kernel as prior knowledge to guide the diffusion model in reconstructing high-resolution CMRI. Furthermore, we propose a cascaded residual attention network feature extractor, which extracts feature information from CMRI low-resolution images for blur kernel prediction and SR reconstruction of CMRI images. Extensive experimental results indicate that our proposed DBSR achieves better blind super-resolution reconstruction results than several state-of-the-art baselines. Our code and model are publicly available at https://github.com/Q-D-F/DBSR .