• About
• Preprints
• Publications
• Abstracts
• Projects
• Education
• Awards

Malte Hoffmann⬏

I am a faculty member in Radiology at Harvard Medical School and Affiliated Faculty in the Health Sciences and Technology Division at MIT. I received a Bachelor’s degree in physics from the University of Paris XI and a Master’s degree and PhD from the University of Cambridge, specializing in the correction of subject motion for brain magnetic resonance imaging (MRI). For my postdoctoral training, I moved to the Massachusetts General Hospital and Harvard Medical School to work with Bruce Fischl, André van der Kouwe, and Adrian Dalca at the Athinoula A Martinos Center for Biomedical Imaging.

My current research focuses on developing algorithms for medical image processing and analysis using artificial intelligence. This work spans innovations in deep learning, computer vision, and MRI acquisition. I am particularly interested in registration, which captures the spatial relationship between objects from images, and in leveraging this information to advance applications with clinical impact, such as fetal MRI.

Preprints⬏

• Anatomy-aware and acquisition-agnostic joint registration with SynthMorph

  Hoffmann M, Hoopes A, Greve DN, Fischl B*, Dalca AV* (*equal contribution)

  arXiv:2301.11329, 2023

Peer-reviewed articles⬏

• Data Consistent Deep Rigid MRI Motion Correction

  Singh NM, Dey N, Hoffmann M, Fischl B, Adalsteinsson E, Frost R*, Dalca AV*, Golland P* (*equal contribution)

  Medical Imaging with Deep Learning (MIDL), forthcoming, 2023

• Equivariant and Denoising CNNs to Decouple Intensity and Spatial Features for Motion Tracking in Fetal Brain MRI

  Billot B, Moyer D, Karani N, Hoffmann M, Abaci Turk E, Grant PE, Golland P

  Medical Imaging with Deep Learning (MIDL), forthcoming, 2023

• Anatomy-specific acquisition-agnostic affine registration learned from fictitious images (PDF)

  Hoffmann M, Hoopes A, Fischl B*, Dalca AV* (*equal contribution)

  SPIE Medical Imaging: Image Processing, 12464, p 1246402, 2023

• SynthStrip: skull-stripping for any brain image

  Hoopes A, Mora JS, Dalca AV, Fischl B*, Hoffmann M* (*equal contribution)

  NeuroImage, 260, p 119474, 2022

• SynthMorph: learning contrast-invariant registration without acquired images

  Hoffmann M, Billot B, Greve DN, Iglesias JE, Fischl B, Dalca AV

  IEEE Transactions on Medical Imaging (TMI), 41 (3), pp 543-558, 2022

• Learning the Effect of Registration Hyperparameters with HyperMorph

  Hoopes A, Hoffmann M, Fischl B, Guttag J, Dalca AV

  Machine Learning for Biomedical Imaging (MELBA), IPMI 2021 Special Issue, 2022

• Probabilistic atlas for the language network based on precision fMRI data from >800 individuals

  Lipkin B, Tuckute G, Affourtit J, Small H, Mineroff Z, Kean H, Jouravlev O, Rakocevic L, Pritchett B, Siegelman M, Hoeflin C, Pongos A, Blank IA, Kline Struhl M, Ivanova A, Shannon S, Sathe A, Hoffmann M, Nieto-Castañón A, Fedorenko E

  Scientific Data, 9, p 529, 2022

• An investigation across 45 languages and 12 language families reveals a universal language network

  Malik-Moraleda S, Ayyash D, Gallée J, Affourtit J, Hoffmann M, Mineroff Z, Jouravlev O, Fedorenko E

  Nature Neuroscience, 25, pp 1014-1019, 2022

• A deep learning toolbox for automatic segmentation of subcortical limbic structures from MRI images

  Greve DN, Billot B, Cordero D, Hoopes A, Hoffmann M, Dalca AV, Fischl B, Iglesias JE, Augustinack JC

  NeuroImage, 244, p 118610, 2021

• Rapid head-pose detection for automated slice prescription of fetal-brain MRI

  Hoffmann M, Abaci Turk E, Gagoski B, Morgan L, Wighton P, Tisdall MD, Reuter M, Adalsteinsson E, Grant PE, Wald LL, van der Kouwe AJW

  International Journal of Imaging Systems and Technology (IMA), 31 (3), pp 1136-1154, 2021

• Learning MRI contrast-agnostic registration

  Hoffmann M, Billot B, Iglesias JE, Fischl B, Dalca AV

  IEEE International Symposium on Biomedical Imaging (ISBI), pp 899-903, 2021

• HyperMorph: Amortized Hyperparameter Learning for Image Registration

  Hoopes A, Hoffmann M, Fischl B, Guttag J, Dalca AV

  Information Processing in Medical Imaging (IPMI), pp 3-17, 2021

• Learning MRI contrast agnostic registration

  Hoffmann M, Billot B, Iglesias JE, Fischl B, Dalca AV

  Medical Imaging Meets NeurIPS (MED-NeurIPS), online only, 2020

• Additional sampling directions improve detection range of wireless radiofrequency probes

  Hoffmann M, Mada M, Carpenter TA, Sawiak SJ, Williams GB

  Magnetic Resonance in Medicine (MRM), 76 (3), pp 913-918, 2016

• Spatial variation of the cooling lines in the reflection nebula NGC 7023

  Bernard-Salas J, Habart E, Köhler M, Abergel A, Arab H, Lebouteiller V, Pinto C, van der Wiel MHD, White GJ, Hoffmann M

  Astronomy & Astrophysics (A&A), 574, pp A97, 2015

• A survey of patient motion in disorders of consciousness and optimization of its retrospective correction

  Hoffmann M, Carpenter TA, Williams GB, Sawiak SJ

  Magnetic Resonance Imaging (MRI), 33 (3), pp 346-350, 2014

Peer-reviewed abstracts⬏

• Can we predict motion artifacts in clinical MRI before the scan completes?

  Hoffmann M, Singh NM, Dalca AV, Fischl B, Frost R

  Annual Meeting of the ISMRM, Toronto, ON, Canada, 2023

• Motion-Aware Neural Networks Improve Rigid Motion Correction of Accelerated Segmented Multislice MRI

  Singh NM, Hoffmann M, Adalsteinsson E, Fischl B, Golland P*, Dalca AV*, Frost R* (*equal contribution)

  Annual Meeting of the ISMRM, Toronto, ON, Canada, 2023

• Clinical evaluation of k-space correlation informed motion artifact detection in segmented multi-slice MRI

  Jang I, Hoffmann M, Singh NM, Balbastre Y, Chen L, Bezerra Cavalcanti Rockenbach MA, Dalca AV, Aganj I, Kalpathy-Cramer J, Fischl B, Frost R

  Annual Meeting of the ISMRM, Toronto, ON, Canada, 2023

• Data Consistency-Based Rigid Motion Correction in Accelerated Multi-Shot 2D-Encoded Brain MRI

  Singh NM, Hoffmann M, Adalsteinsson E, Fischl B, Golland P, Dalca AV, Frost R

  ISMRM Workshop on Motion Detection & Correction, Oxford, United Kingdom, 2022

• SynthStrip: skull stripping for any brain image

  Hoopes A, Mora JS, Dalca AV, Fischl B*, Hoffmann M* (*equal contribution)

  Annual Meeting of the ISMRM, London, United Kingdom, 2022

• Joint Neural Network for Fast Retrospective Rigid Motion Correction of Accelerated Segmented Multislice MRI

  Singh NM, Hoffmann M, Moyer DC, Jang I, Chen L, Bezerra Cavalcanti Rockenbach MA, Guidon A, Aganj I, Kalpathy-Cramer J, Adalsteinsson E, Fischl B, Dalca AV, Golland P*, Frost R* (*equal contribution)

  Annual Meeting of the ISMRM, London, United Kingdom, 2022

• Automated MRI k-space Motion Artifact Detection in Segmented Multi-Slice Sequences

  Jang I*, Frost R*, Hoffmann M, Singh NM, Chen L, Guidon A, Bezerra Cavalcanti Rockenbach MA, Comeau D, Bizzo B, Chang K, Witham S, Rettmann D, Banerjee S, Brau A, Reese T, Aganj I, Dalca AV, Fischl B*, Kalpathy-Cramer J* (*equal contribution)

  Annual Meeting of the ISMRM, London, United Kingdom, 2022

• Learning-based automatic field-of-view positioning for fetal-brain MRI

  Hoffmann M, Moyer DC, Zhang L, Golland P, Grant PE, Gagoski B, van der Kouwe AJW

  Annual Meeting of the ISMRM, 2021

• Learning-based non-linear registration robust to MRI-sequence contrast

  Hoffmann M, Billot B, Iglesias JE, Fischl B, Dalca AV

  Annual Meeting of the ISMRM, 2021

• Longitudinal FreeSurfer with non-linear subject-specific template improves sensitivity to cortical thinning

  Hoffmann M, Salat DH, Reuter M*, Fischl B* (*equal contribution)

  Annual Meeting of the ISMRM, 2020

• Real-time brain masking algorithm improves motion tracking accuracy in scans with volumetric navigators (vNavs)

  Hoffmann M, Frost R, Salat DH, Tisdall MD, Polimeni J, van der Kouwe AJW

  Annual Meeting of the ISMRM, 2020

• Scan-specific assessment of vNav motion artifact mitigation in the HCP Aging study using reverse motion correction

  Frost R, Tisdall MD, Hoffmann M, Fischl B, Salat DH, van der Kouwe AJW

  Annual Meeting of the ISMRM, 2020

• Improved, rapid fetal-brain localization and orientation detection for auto-slice prescription

  Hoffmann M, Gagoski B, Abaci Turk E, Wighton P, Tisdall MD, Reuter M, Adalsteinsson E, Grant PE, Wald LL, van der Kouwe AJW

  Annual Meeting of the ISMRM, Montreal, QC, Canada, 2019

• Fast, automated slice prescription of standard anatomical planes for fetal brain MRI

  Hoffmann M, Gagoski B, Abaci Turk E, Wighton P, Tisdall MD, Reuter M, Adalsteinsson E, Grant PE, Wald LL, van der Kouwe AJW

  Annual Meeting of the ISMRM, Paris, France, 2018

• Evaluation of prospective and retrospective motion correction exploiting the slice-based acquisition of fMRI

  Hoffmann M, Sawiak SJ

  Annual Meeting of the ISMRM, Honolulu, HI, United States, 2017

• Automated rejection of motion-corrupted slices and optimised retrospective ghost correction for multi-shot DTI

  Hoffmann M, Sawiak SJ

  Annual Meeting of the ISMRM, Singapore, 2016

• Should volumetric, slice-wise or non-linear registration be used for motion correction of fMRI data?

  Hoffmann M, Sawiak SJ

  Annual Meeting of the ISMRM, Singapore, 2016

• Additional sampling directions improve detection range of wireless RF probes for use in non-compliant patients

  Hoffmann M, Mada M, Carpenter TA, Sawiak SJ, Williams GB

  ISMRM British Chapter Annual Postgraduate Meeting, Stevenage, United Kingdom, 2015

• Optimisation of retrospective motion correction in fMRI of uncooperative patients

  Hoffmann M, Carpenter TA, Williams GB, Sawiak SJ

  ISMRM British Chapter Annual Scientific Meeting, Edinburgh, United Kingdom, 2014

• A survey of patient motion in disorders of consciousness and optimisation of its retrospective correction

  Hoffmann M, Carpenter TA, Williams GB, Sawiak SJ

  ISMRM Workshop on Motion Correction in MRI, Tromsø, Norway, 2014

Projects⬏

• SynthStrip

  SynthStrip is a robust skull-stripping tool that removes non-brain image content from full-head scans across imaging modalities, contrasts, resolutions, and subject populations. The SynthStrip dataset is a freely available collection of full-head images with corresponding brain masks from 622 MRI, CT, and PET scans.

• SynthMorph

  SynthMorph, a strategy for learning contrast-invariant registration without acquired images, enables unprecedented network generalization across a landscape of real image types. By exposing deep neural networks to wildly varying synthetic training data, SynthMorph alleviates the need for retraining models.

• Fetal-Align

  Fetal-Align is a fetal-head pose detection algorithm for full-uterus stacks of echo planar imaging (EPI) slices. The project aims at improving fetal MRI by automating the clinical workflow and includes a small testset of 41 EPI stacks from fetuses at 26-37 weeks' gestation.

Education⬏

• 2017-2021 Postdoctoral Fellow, Radiology

  Harvard Medical School, United States

  Massachusetts General Hospital, United States

• 2013-2017 Doctor of Philosophy (PhD), MRI Physics

  University of Cambridge, United Kingdom

• 2011-2012 Master of Advanced Studies (MASt), Experimental and Theoretical Physics

  University of Cambridge, United Kingdom

• 2008-2011 Selective Bachelor's degree (Licence et Magistère 1), Fundamental Physics

  University of Paris XI, France

Awards⬏

• 2020, 2021 ISMRM Summa Cum Laude Merit Award for top 5% abstract

• 2020 NIH K99/R00 Pathway to Independence Award

• 2019 ISMRM Detection & Correction of Motion in MRI & MRS Study Group Award

• 2016-2019 ISMRM Educational Stipend

• 2015, 2016 Cusanuswerk Graduate Travel Award

• 2014, 2015 Downing College Graduate Travel Fund Award

• 2014 University of Cambridge Medical School EG Fearnside Travel Fund Award

• 2014 ISMRM Workshop on Motion Correction in MRI Trainee Stipend

• 2013-2016 Cusanuswerk Graduate Scholarship

• 2010-2012 Cusanuswerk Undergraduate Scholarship

• 2010 University of Paris XI IT Grant Award