A survey of emerging applications of diffusion probabilistic models in MRI

Diffusion probabilistic models (DPMs) which employ explicit likelihood characterization and a gradual sampling process to synthesize data, have gained increasing research interest. Despite their huge computational burdens due to the large number of steps involved during sampling, DPMs are widely app...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuheng Fan, Hanxi Liao, Shiqi Huang, Yimin Luo, Huazhu Fu, Haikun Qi
Format: Article
Language:English
Published: KeAi Communications Co., Ltd. 2024-06-01
Series:Meta-Radiology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2950162824000353
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846170054417186816
author Yuheng Fan
Hanxi Liao
Shiqi Huang
Yimin Luo
Huazhu Fu
Haikun Qi
author_facet Yuheng Fan
Hanxi Liao
Shiqi Huang
Yimin Luo
Huazhu Fu
Haikun Qi
author_sort Yuheng Fan
collection DOAJ
description Diffusion probabilistic models (DPMs) which employ explicit likelihood characterization and a gradual sampling process to synthesize data, have gained increasing research interest. Despite their huge computational burdens due to the large number of steps involved during sampling, DPMs are widely appreciated in various medical imaging tasks for their high-quality and diversity of generation. Magnetic resonance imaging (MRI) is an important medical imaging modality with excellent soft tissue contrast and superb spatial resolution, which possesses unique opportunities for DPMs. Although there is a recent surge of studies exploring DPMs in MRI, a survey paper of DPMs specifically designed for MRI applications is still lacking. This review article aims to help researchers in the MRI community to grasp the advances of DPMs in different applications. We first introduce the theory of two dominant kinds of DPMs, categorized according to whether the diffusion time step is discrete or continuous, and then provide a comprehensive review of emerging DPMs in MRI, including reconstruction, image generation, image translation, segmentation, anomaly detection, and further research topics. Finally, we discuss the general limitations as well as limitations specific to the MRI tasks of DPMs and point out potential areas that are worth further exploration.
format Article
id doaj-art-eccced6f2e544386b77ec5c3f8c3b42b
institution Kabale University
issn 2950-1628
language English
publishDate 2024-06-01
publisher KeAi Communications Co., Ltd.
record_format Article
series Meta-Radiology
spelling doaj-art-eccced6f2e544386b77ec5c3f8c3b42b2024-11-12T05:22:43ZengKeAi Communications Co., Ltd.Meta-Radiology2950-16282024-06-0122100082A survey of emerging applications of diffusion probabilistic models in MRIYuheng Fan0Hanxi Liao1Shiqi Huang2Yimin Luo3Huazhu Fu4Haikun Qi5School of Biomedical Engineering, ShanghaiTech University, Pudong, Shanghai, 201210, ChinaSchool of Biomedical Engineering, ShanghaiTech University, Pudong, Shanghai, 201210, ChinaSchool of Biomedical Engineering, ShanghaiTech University, Pudong, Shanghai, 201210, ChinaSchool of Biomedical Engineering and Imaging Sciences, King's College London, St. Thomas' Hospital, London, SE1 7EH, UKInstitute of High Performance Computing, Agency for Science, Technology, and Research (A∗STAR), 201210, SingaporeSchool of Biomedical Engineering, ShanghaiTech University, Pudong, Shanghai, 201210, China; Corresponding author.Diffusion probabilistic models (DPMs) which employ explicit likelihood characterization and a gradual sampling process to synthesize data, have gained increasing research interest. Despite their huge computational burdens due to the large number of steps involved during sampling, DPMs are widely appreciated in various medical imaging tasks for their high-quality and diversity of generation. Magnetic resonance imaging (MRI) is an important medical imaging modality with excellent soft tissue contrast and superb spatial resolution, which possesses unique opportunities for DPMs. Although there is a recent surge of studies exploring DPMs in MRI, a survey paper of DPMs specifically designed for MRI applications is still lacking. This review article aims to help researchers in the MRI community to grasp the advances of DPMs in different applications. We first introduce the theory of two dominant kinds of DPMs, categorized according to whether the diffusion time step is discrete or continuous, and then provide a comprehensive review of emerging DPMs in MRI, including reconstruction, image generation, image translation, segmentation, anomaly detection, and further research topics. Finally, we discuss the general limitations as well as limitations specific to the MRI tasks of DPMs and point out potential areas that are worth further exploration.http://www.sciencedirect.com/science/article/pii/S2950162824000353Diffusion probabilistic modelsScore based generative modelingMRI
spellingShingle Yuheng Fan
Hanxi Liao
Shiqi Huang
Yimin Luo
Huazhu Fu
Haikun Qi
A survey of emerging applications of diffusion probabilistic models in MRI
Meta-Radiology
Diffusion probabilistic models
Score based generative modeling
MRI
title A survey of emerging applications of diffusion probabilistic models in MRI
title_full A survey of emerging applications of diffusion probabilistic models in MRI
title_fullStr A survey of emerging applications of diffusion probabilistic models in MRI
title_full_unstemmed A survey of emerging applications of diffusion probabilistic models in MRI
title_short A survey of emerging applications of diffusion probabilistic models in MRI
title_sort survey of emerging applications of diffusion probabilistic models in mri
topic Diffusion probabilistic models
Score based generative modeling
MRI
url http://www.sciencedirect.com/science/article/pii/S2950162824000353
work_keys_str_mv AT yuhengfan asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT hanxiliao asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT shiqihuang asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT yiminluo asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT huazhufu asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT haikunqi asurveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT yuhengfan surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT hanxiliao surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT shiqihuang surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT yiminluo surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT huazhufu surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri
AT haikunqi surveyofemergingapplicationsofdiffusionprobabilisticmodelsinmri