Fang Liu
Fang Liu is a radiology researcher whose work applies machine learning to magnetic resonance imaging. She coauthored research on quantitative relaxometry, deep-learning PET/MR correction, the MANTIS reconstruction framework, adversarial MRI mapping, and the self-supervised RELAX method. In 2025, she coauthored a meta-learning approach to multi-task MRI reconstruction. As of October 2026, she is an associate professor of radiology at the Martinos Center for Biomedical Imaging.
- Now at
- Mass General Brigham and Harvard University
- Works
- 124
- Citations
- 4,621
- h-index
- 29
Publication numbers from OpenAlex, as of 27 October 2026.
Timeline
-
2015
Correcting relaxometry bias
In 2015, she coauthors mcRISE, a method that corrects magnetization-transfer bias in multicomponent relaxometry. She is affiliated with the University of Wisconsin School of Medicine and Public Health's Department of Medical Physics.
-
2017
Deep MRAC for PET/MR
In 2017, she coauthors deep MRAC, a deep-learning approach that generates pseudo-CT scans from MRI for PET/MR imaging. Its reconstruction errors average below 1% in most brain regions.
-
2018
MANTIS for MR mapping
In 2018, she coauthors MANTIS, a model-augmented neural network that combines end-to-end CNN mapping, incoherent k-space sampling, and a physical model to reconstruct knee T2 maps.
-
2020
Adversarial MRI reconstruction
In 2020, she and colleagues at MGH's Gordon Center for Medical Imaging add a generative adversarial network to MANTIS to accelerate image reconstruction. Their feasibility study reports greater accuracy and sharper image texture than conventional methods.
-
2021
The RELAX framework
In 2021, she and colleagues publish RELAX, a model-guided self-supervised deep-learning framework that reconstructs quantitative MR parameter maps from undersampled k-space.
-
2025
Meta-learning for MRI reconstruction
In 2025, she coauthors a meta-learning method for multi-task MRI reconstruction.
-
As of October 2026
Associate professor of radiology
As of October 2026, she is an associate professor of radiology at the Martinos Center for Biomedical Imaging.
Sources and credits
Photo: We have no published photo of Fang Liu yet. Send us a free-licensed photo.
- Rapid Multi-Component Relaxometry in Steady State with Correction of Magnetization Transfer Effects - PMC
- Deep Learning MR Imaging–based Attenuation Correction for PET/MR Imaging - PMC
- [1809.03308] MANTIS: Model-Augmented Neural neTwork with Incoherent k-space Sampling for efficient MR T2 mapping
- Artificial Intelligence Accelerates MRI Parameter Mapping - Mass General Advances in Motion
- Magnetic resonance parameter mapping using model-guided self-supervised deep learning - PMC
- Liu, Fang – Martinos Center