Malte Hoffmann
Assistant Professor
Jingru Fu
Postdoctoral Fellow
Yining Jiao
Postdoctoral Fellow
Junho Moon
Visiting PhD Student
Assistant Professor of Radiology, Harvard Medical School
Investigator, Massachusetts General Hospital
Brain imaging AI that works on any scan, out of the box
We develop deep-learning methods for brain segmentation and registration, which delineate anatomy and align it across images. Imaging AI often fails on unseen image types. We train models on wildly varied synthetic data so they work on new image types without retraining. Our tools are standard components of FreeSurfer, widely used brain MRI software developed at the Martinos Center for Biomedical Imaging.
I have advised companies building imaging AI on model robustness, segmentation, and registration. We also welcome sponsored research partnerships, for example to adapt our tools to new applications.
Skull-stripping for any brain image, with a dataset of brain masks for 622 MRI, CT, and PET scans
Rigid, affine, and deformable brain registration across scan types, without preprocessing
Brain segmentation for infants and young children from one or more MRIs of variable resolution and type
A selection of our work, starting with a practical guide to the idea behind our tools: training networks on synthetic data so they work on any brain scan. A complete list is on Google Scholar. Asterisks indicate equal contribution.
Domain-randomized deep learning for neuroimage analysis (arXiv, PMC)
Hoffmann M
IEEE Signal Processing Magazine (SPM), 42 (4), pp 78-90, 2025
SynthMorph: learning contrast-invariant registration without acquired images (arXiv, PMC)
Hoffmann M, Billot B, Greve DN, Iglesias JE, Fischl B, Dalca AV
IEEE Transactions on Medical Imaging (TMI), 41 (3), pp 543-558, 2022
Anatomy-aware and acquisition-agnostic joint registration with SynthMorph (arXiv, PMC)
Hoffmann M, Hoopes A, Greve DN, Fischl B*, Dalca AV*
Imaging Neuroscience, 2, pp 1-33, 2024
SynthStrip: skull-stripping for any brain image (arXiv, PMC)
Hoopes A, Mora JS, Dalca AV, Fischl B*, Hoffmann M*
NeuroImage, 260, p 119474, 2022
Deep infant brain segmentation from multi-contrast MRI (arXiv, PMC)
Hoffmann M, Zöllei L*, Dalca AV*
Asilomar Conference on Signals, Systems, and Computers (ACSSC), pp 974-981, 2026
DisMorph: learning to disentangle technical distortions from true biological change (arXiv)
Fu J, Larson KE, Greve DN, Fischl B*, Hoffmann M*
MICCAI Workshop on Simulation and Synthesis in Medical Imaging (SASHIMI), in press, 2026
Malte Hoffmann
Assistant Professor
Jingru Fu
Postdoctoral Fellow
Yining Jiao
Postdoctoral Fellow
Junho Moon
Visiting PhD Student
2026 Visiting High School Student, Choate Rosemary Hall
2023-2025 Remote Postdoctoral Researcher
Next: Assistant Professor, University of Haifa
2023-2024 Postdoctoral Fellow
Next: Postdoctoral Researcher, University of Pennsylvania
2023-2024 Research Technician, co-supervised with Lilla Zöllei
Next: Data Analyst, Mass General Brigham
2022 Visiting MSc Student, Northeastern University
Next: PhD Student, Indiana University Indianapolis
We have hosted graduate students and postdocs leading projects in our group, with support from programs such as WASP or DAAD. To inquire about a stay of at least six months, ideally a year, email a CV and a sentence on what you would like to work on.
2026 NIH R21 Trailblazer Award for New and Early Stage Investigators
2020 NIH K99/R00 Pathway to Independence Award
2020, 2021 ISMRM Summa Cum Laude Merit Awards for top-5% abstracts
2019 ISMRM Detection & Correction of Motion in MRI & MRS Study Group Award
2013-2016 Cusanuswerk Graduate Scholarship
2017-2022 Postdoctoral Fellow, Radiology (with van der Kouwe, Fischl, Dalca)
Harvard Medical School, USA
Massachusetts General Hospital, USA
2013-2017 Doctor of Philosophy (PhD), Motion Correction in MRI
University of Cambridge, UK
2011-2012 Master of Advanced Studies (MASt), Experimental and Theoretical Physics
University of Cambridge, UK
2008-2011 Bachelor's Degree (Licence et Magistère 1), Fundamental Physics
University of Paris XI, France
Athinoula A. Martinos Center for Biomedical Imaging
149 13th Street, Suite 2301 // 10071
Charlestown, MA 02129
United States
