Overview
Background
Dr Hongfu Sun completed his PhD in Biomedical Engineering at the University of Alberta in 2015, followed by postdoctoral training in Calgary until 2018. He joined the Imaging, Sensing and Biomedical Engineering team in the School of ITEE at UQ in 2019 and was awarded the ARC DECRA fellowship in 2021. His research interests include developing novel magnetic resonance imaging (MRI) contrast mechanisms, e.g. Quantitative Susceptibility Mapping (QSM), fast and multi-parametric MRI acquisitions, and advanced image reconstruction techniques, including deep learning and artificial intelligence, to advance medical imaging techniques for clinical applications.
Dr Sun is currently recruiting graduate students. Check out Available Projects for details. Open to both Domestic and International students.
Availability
- Dr Hongfu Sun is:
- Available for supervision
Fields of research
Qualifications
- Doctor of Philosophy, University of Alberta
Research interests
-
MR image processing through advanced optimization techniques and deep learning
My research interest is to develop advanced image reconstruction and processing methods to solve some of the mathematical challenges in MRI research, such as the ill-posed inverse problem in Quantitative Susceptibility Mapping (QSM). Some of the approaches I am particularly interested in are (1) image optimization techniques such as image regularization and compressed sensing, and (2) machine learning and especially deep learning through the convolutional neural network (CNN) and transformers in vision.
-
Fast, multi-parametric, and quantitative MRI acquisition methods at ultra-high field
A single MRI acquisition usually takes 3-5 minutes, and a standard clinical protocol may require a couple of them for complementary contrasts. My research interest is to significantly reduce the total scan time for each patient while maintaining the same amount of information for diagnosis, by (1) accelerating individual scans with parallel imaging techniques at ultra-high field, and (2) designing novel MRI sequences that can produce multi-contrast weighted images and multi-parametric quantitative maps from a single MR acquisition.
-
Brain imaging applications in neuroscience and neurological diseases
MRI is one of the best tools to study the brain in vivo, thanks to its excellent soft tissue contrast and its versatile contrast mechanisms. I am interested in applying advanced and comprehensive image analysis on different MRI methods to better understand neuroscience, such as brain development in children and structural and functional connectivity of the brain, as well as some of the neurological diseases such as Alzheimer/Dementia, Multiple Sclerosis, Schizophrenia, and Stroke.
Research impacts
Dr Hongfu Sun is one of the early pioneers in developing a novel MRI technique - Quantitative Susceptibility Mapping (QSM), which is one of the most significant MRI contrast breakthroughs in recent years, that has demonstrated wide clinical applications in healthy, aging and diseased human brains, such as dementia, Alzheimer's disease, Parkinson's disease, multiple sclerosis, schizophrenia, stroke, etc. Since commencing at UQ, Dr Sun has extended his research topics to exploiting novel reconstruction algorithms using state-of-the-art deep learning-based artificial intelligence techniques.
Works
Search Professor Hongfu Sun’s works on UQ eSpace
2024
Conference Publication
QSM Reconstruction of Arbitrary Dipole Orientations using an End-to-end Neural Network via Latent Feature Editing
Gao, Yang, Xiong, Zhuang, Shan, Shanshan, Li, Min, Wilman, Alan H, Pike, G. Bruce, Liu, Feng and Sun, Hongfu (2024). QSM Reconstruction of Arbitrary Dipole Orientations using an End-to-end Neural Network via Latent Feature Editing. 2024 ISMRM & ISMRT Annual Meeting, Singapore, 4-9 May 2024. Concord, CA United States: ISMRM. doi: 10.58530/2024/2453
2023
Journal Article
Editorial: Imaging of neurometabolism
Zheng, Wenbin, Dai, Zhouzhi, Wu, Renhua and Sun, Hongfu (2023). Editorial: Imaging of neurometabolism. Frontiers in Neuroscience, 17 1286361, 1286361. doi: 10.3389/fnins.2023.1286361
2023
Journal Article
MapFlow: latent transition via normalizing flow for unsupervised domain adaptation
Askari, Hossein, Latif, Yasir and Sun, Hongfu (2023). MapFlow: latent transition via normalizing flow for unsupervised domain adaptation. Machine Learning, 112 (8), 2953-2974. doi: 10.1007/s10994-023-06357-2
2023
Journal Article
Increased glymphatic system activity in patients with mild traumatic brain injury
Dai, Zhuozhi, Yang, Zhiqi, Li, Zhaolin, Li, Mu, Sun, Hongfu, Zhuang, Zerui, Yang, Weichao, Hu, Zehuan, Chen, Xiaofeng, Lin, Daiying and Wu, Xianheng (2023). Increased glymphatic system activity in patients with mild traumatic brain injury. Frontiers in Neurology, 14 1148878, 1-7. doi: 10.3389/fneur.2023.1148878
2023
Journal Article
Distortion‐corrected image reconstruction with deep learning on an MRI‐Linac
Shan, Shanshan, Gao, Yang, Liu, Paul Z. Y., Whelan, Brendan, Sun, Hongfu, Dong, Bin, Liu, Feng and Waddington, David E. J. (2023). Distortion‐corrected image reconstruction with deep learning on an MRI‐Linac. Magnetic Resonance in Medicine, 90 (3), 1-15. doi: 10.1002/mrm.29684
2023
Journal Article
Affine transformation edited and refined deep neural network for quantitative susceptibility mapping
Xiong, Zhuang, Gao, Yang, Liu, Feng and Sun, Hongfu (2023). Affine transformation edited and refined deep neural network for quantitative susceptibility mapping. NeuroImage, 267 119842, 1-9. doi: 10.1016/j.neuroimage.2022.119842
2023
Conference Publication
QSM from the raw phase using an end-to-end neural network
Gao, Yang, Xiong, Zhuang, Fazlollahi, Amir, Nestor, Peter, Vegh, Viktor, Pike, G. Bruce, Crozier, Stuart, Liu, Feng and Sun, Hongfu (2023). QSM from the raw phase using an end-to-end neural network. ISMRM Annual Meeting, London, United Kingdom, 7-12 May 2022. Berkeley, CA, United States: International Society for Magnetic Resonance in Medicine. doi: 10.58530/2022/4740
2022
Journal Article
Quantitative susceptibility mapping changes relate to gait issues in Parkinson’s Disease
Nathoo, Nabeela, Gee, Myrlene, Nelles, Krista, Burt, Jacqueline, Sun, Hongfu, Seres, Peter, Wilman, Alan H., Beaulieu, Christian, Ba, Fang and Camicioli, Richard (2022). Quantitative susceptibility mapping changes relate to gait issues in Parkinson’s Disease. Canadian Journal of Neurological Sciences, 50 (6) PII S031716712200316X, 1-8. doi: 10.1017/cjn.2022.316
2022
Journal Article
Detecting monocyte trafficking in an animal model of glioblastoma using R2* and quantitative susceptibility mapping
Yang, Runze, Hamilton, A. Max, Sun, Hongfu, Rawji, Khalil S., Sarkar, Susobhan, Mirzaei, Reza, Pike, G. Bruce, Yong, V. Wee. and Dunn, Jeff F. (2022). Detecting monocyte trafficking in an animal model of glioblastoma using R2* and quantitative susceptibility mapping. Cancer Immunology, Immunotherapy, 72 (3), 733-742. doi: 10.1007/s00262-022-03297-z
2022
Journal Article
Brain volume and magnetic susceptibility differences in children and adolescents with prenatal alcohol exposure
Nakhid, Daphne, McMorris, Carly, Sun, Hongfu, Gibbard, William Benton, Tortorelli, Christina and Lebel, Catherine (2022). Brain volume and magnetic susceptibility differences in children and adolescents with prenatal alcohol exposure. Alcoholism: Clinical and Experimental Research, 46 (10), 1797-1807. doi: 10.1111/acer.14928
2022
Journal Article
Quantitative susceptibility-weighted imaging in presence of strong susceptibility sources: application to hemorrhage
De, Ashmita, Sun, Hongfu, Emery, Derek J., Butcher, Kenneth S. and Wilman, Alan H. (2022). Quantitative susceptibility-weighted imaging in presence of strong susceptibility sources: application to hemorrhage. Magnetic Resonance Imaging, 92, 224-231. doi: 10.1016/j.mri.2022.06.010
2022
Journal Article
Instant tissue field and magnetic susceptibility mapping from MRI raw phase using Laplacian enhanced deep neural networks
Gao, Yang, Xiong, Zhuang, Fazlollahi, Amir, Nestor, Peter J., Vegh, Viktor, Nasrallah, Fatima, Winter, Craig, Pike, G. Bruce, Crozier, Stuart, Liu, Feng and Sun, Hongfu (2022). Instant tissue field and magnetic susceptibility mapping from MRI raw phase using Laplacian enhanced deep neural networks. NeuroImage, 259 119410, 1-13. doi: 10.1016/j.neuroimage.2022.119410
2022
Journal Article
BFRnet: A deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources
Zhu, Xuanyu, Gao, Yang, Liu, Feng, Crozier, Stuart and Sun, Hongfu (2022). BFRnet: A deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources. Zeitschrift fur Medizinische Physik, 33 (4), 578-590. doi: 10.1016/j.zemedi.2022.08.001
2022
Journal Article
Brain iron and mental health symptoms in youth with and without prenatal alcohol exposure
Nakhid, Daphne, McMorris, Carly A., Sun, Hongfu, Gibbard, Ben, Tortorelli, Christina and Lebel, Catherine (2022). Brain iron and mental health symptoms in youth with and without prenatal alcohol exposure. Nutrients, 14 (11) 2213, 1-18. doi: 10.3390/nu14112213
2022
Journal Article
Distinguishing COVID-19 from influenza pneumonia in the early stage through CT imaging and clinical features
Yang, Zhiqi, Lin, Daiying, Chen, Xiaofeng, Qiu, Jinming, Li, Shengkai, Huang, Ruibin, Yang, Zhijian, Sun, Hongfu, Liao, Yuting, Xiao, Jianning, Tang, Yanyan, Chen, Xiangguang, Zhang, Sheng and Dai, Zhuozhi (2022). Distinguishing COVID-19 from influenza pneumonia in the early stage through CT imaging and clinical features. Frontiers in Microbiology, 13 847836, 1-9. doi: 10.3389/fmicb.2022.847836
2022
Journal Article
Deep grey matter quantitative susceptibility mapping from small spatial coverages using deep learning
Zhu, Xuanyu, Gao, Yang, Liu, Feng, Crozier, Stuart and Sun, Hongfu (2022). Deep grey matter quantitative susceptibility mapping from small spatial coverages using deep learning. Zeitschrift fur Medizinische Physik, 32 (2), 188-198. doi: 10.1016/j.zemedi.2021.06.004
2021
Journal Article
Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction
Gao, Yang, Cloos, Martijn, Liu, Feng, Crozier, Stuart, Pike, G. Bruce and Sun, Hongfu (2021). Accelerating quantitative susceptibility and R2* mapping using incoherent undersampling and deep neural network reconstruction. NeuroImage, 240 118404, 1-13. doi: 10.1016/j.neuroimage.2021.118404
2021
Journal Article
Inter-observer variations of the tumor bed delineation for patients after breast conserving surgery in preoperative magnetic resonance and computed tomography scan fusion
Jiang, Jie, Chen, Jinhu, Li, Wanhu, Li, Yongqing, Chen, Yiru, Zhang, Zicheng, Liu, Chengxin, Han, Dan, Sun, Hongfu, Li, Baosheng and Huang, Wei (2021). Inter-observer variations of the tumor bed delineation for patients after breast conserving surgery in preoperative magnetic resonance and computed tomography scan fusion. Bmc Cancer, 21 (1) 838. doi: 10.1186/s12885-021-08546-5
2021
Conference Publication
Accelerating QSM using compressed sensing and deep neural network
Gao, Yang, Liu, Feng, Crozier, Stuart and Sun, Hongfu (2021). Accelerating QSM using compressed sensing and deep neural network. 2021 ISMRM & SMRT Annual Meeting & Exhibition, Online, 15-20 May 2021. Berkeley, CA, United States: International Society for Magnetic Resonance in Medicine.
2021
Journal Article
On the regularization of feature fusion and mapping for fast MR multi-contrast imaging via iterative networks
Liu, Xinwen, Wang, Jing, Sun, Hongfu, Chandra, Shekhar S, Crozier, Stuart and Liu, Feng (2021). On the regularization of feature fusion and mapping for fast MR multi-contrast imaging via iterative networks. Magnetic resonance imaging, 77, 159-168. doi: 10.1016/j.mri.2020.12.019
Funding
Current funding
Past funding
Supervision
Availability
- Dr Hongfu Sun is:
- Available for supervision
Looking for a supervisor? Read our advice on how to choose a supervisor.
Available projects
-
MRI and deep learning methods development and applications at ultra-high field
I am currently recruiting Master and PhD students to innovate on novel MRI methods and deep learning image reconstruction techniques that can be eventually applied to neuroscience and neurological diseases. We have an excellent and accessible MRI facility here at UQ, e.g. a state-of-the-art 3T Prisma and a prestigious 7T whole-body system (only two in Australia, the other one in UniMelb). The research projects will involve MRI physics, pulse sequence programming, image processing (e.g. deep learning), and image analysis. By the end of your graduate study, you will be an expert in MRI with comprehensive skills in maths, physics, computer programming, and artificial intelligence.
https://graduate-school.uq.edu.au/project/developing-ai-based-mri-methods-microscopic-imaging
Supervision history
Current supervision
-
Doctor Philosophy
MR image processing through advanced optimisation techniques and deep learning
Principal Advisor
Other advisors: Professor Feng Liu
-
Doctor Philosophy
MR image processing through advanced optimisation techniques and deep learning
Principal Advisor
Other advisors: Professor Feng Liu
-
Doctor Philosophy
Development of novel deep learning methods for medical imaging
Principal Advisor
Other advisors: Professor Feng Liu, Dr Nan Ye
-
Doctor Philosophy
MRI methods development through deep learning
Principal Advisor
Other advisors: Professor Feng Liu
-
Doctor Philosophy
Combined Compressed sensing and machine learning/deep learning methods for rapid MRI
Associate Advisor
Other advisors: Professor Kwun Fong, Associate Professor Henry Marshall, Professor Feng Liu
-
Doctor Philosophy
Magnetic Resonance Image Processing with Artificial Intelligence
Associate Advisor
Other advisors: Associate Professor Craig Engstrom, Dr Shakes Chandra
Completed supervision
-
2025
Doctor Philosophy
Robust and generalizable deep Learning quantitative susceptibility mapping for human brains
Principal Advisor
Other advisors: Professor Feng Liu
-
2024
Doctor Philosophy
Key Applications in Deep Learning Based Quantitative Susceptibility Mapping
Associate Advisor
Other advisors: Emeritus Professor Stuart Crozier, Professor Feng Liu
-
2022
Doctor Philosophy
Deep Learning-based Quantitative Susceptibility Mapping: Methods Development and Applications
Associate Advisor
Other advisors: Emeritus Professor Stuart Crozier, Professor Feng Liu
Media
Enquiries
For media enquiries about Dr Hongfu Sun's areas of expertise, story ideas and help finding experts, contact our Media team: