Cine Magnetic Resonance
Cine magnetic resonance imaging (cine MRI) provides dynamic visualization of the heart, crucial for assessing cardiac function and diagnosing diseases. Current research heavily focuses on accelerating cine MRI acquisition through deep learning techniques, employing convolutional neural networks, recurrent networks, and diffusion models to reconstruct high-quality images from undersampled data, often incorporating advanced strategies like attention mechanisms and paired sampling to improve image sharpness and reduce artifacts. These advancements aim to reduce scan times, enabling single breath-hold imaging and free-breathing acquisitions, ultimately improving patient comfort and diagnostic accuracy. The resulting improvements in image quality and efficiency have significant implications for clinical practice and large-scale epidemiological studies.
Papers
Spatiotemporal Diffusion Model with Paired Sampling for Accelerated Cardiac Cine MRI
Shihan Qiu, Shaoyan Pan, Yikang Liu, Lin Zhao, Jian Xu, Qi Liu, Terrence Chen, Eric Z. Chen, Xiao Chen, Shanhui Sun
Clinically Feasible Diffusion Reconstruction for Highly-Accelerated Cardiac Cine MRI
Shihan Qiu, Shaoyan Pan, Yikang Liu, Lin Zhao, Jian Xu, Qi Liu, Terrence Chen, Eric Z. Chen, Xiao Chen, Shanhui Sun