News

← Back to home

2026

  • November 2026: A free online workshop by the FAIMI initiative is planned for 24 November, with a keynote by Dr. Ghada Zamzmi (HeartFlow) on "Bridging Regulatory Science and Medical Imaging Evaluation".
  • October 2026: Moderating the Oxford-style debate "Should AI agents replace PhD students in research projects?" (Anna Rogers vs. Sune Lehmann) at the D3A Conference, Hotel Nyborg Strand, 9 October – as a member of the conference's scientific programme committee.
  • September 2026: MICCAI 2026 week for the lab – the FAIMI workshop ran for the fourth year running, with a keynote by Aasa Feragen; Llucia Coll Benejam presented an oral at the MICCAI Open Data Day introducing a new clinical polymicrogyria brain MRI dataset (published in MELBA); the same dataset featured as a task in the 2026 FOMO foundation-model challenge, with Llucia Coll Benejam and Alice Schiavone leading the lab's participation.
  • September 2026: Congratulations to TRUSTMIND postdoc Alissa Valentine on her first first-author postdoc paper, showing that AI models predicting depression diagnoses from Danish registry data perform unevenly across immigration status, socioeconomic background, and region. She will present it at the Responsible AI for Mental Health Workshop (6 October), the D3A Conference (8 October), ECNP Congress (11 October) and the AIES Conference (13 October).
  • September 2026: New paper in the Journal of Cerebral Blood Flow & Metabolism, "Building an open infrastructure for molecular neuroimaging: standards and tools from the OpenNeuroPET initiative" (with Martin Nørgaard, Gitte Moos Knudsen, Russell Poldrack, Cyril Pernet, Granville Matheson and Vince Calhoun), bringing together PET-BIDS, OpenNeuro/PublicnEUro data sharing, and reproducible workflow tools including PETPrep (led by Martin Nørgaard).
  • July 2026: Postdoc Madeleine Wyburd presented the first preliminary data from the NIBS-CP cohort at the FITNG2026 meeting, led by PI Kathrine Skak Madsen and Co-PI Melanie Ganz. Recruitment for the cohort is expected to complete in autumn 2026.
  • June 2026: The LNN@UCPH attended NRM2026 in Uppsala with three oral presentations involving lab members.
  • June 2026: The LNN@UCPH was present in force at OHBM 2026 in Bordeaux – four abstracts accepted, and involvement in two educational sessions, including one on open science for infant MRI (14 June), co-organised with Madeleine Wyburd, Dafnis Batallé, Elise Turk, Brittany Howell, Lilla Zollei and Eric Feczko.
  • May 2026: Presented the poster "Large sample evaluation of AI accelerated Clinical Pediatric Neuroimaging" (Karen J. Kettless, Camilla D. Hansen, Llucia Coll, Marcel Dominik Nickel, Adam E. Hansen, Melanie Ganz) at ISMRM 2026 in Cape Town, and helped run the ISMRM Motion Correction study group meeting as part of its leadership team (with Adam van Niekerk, Thomas Küstner and Stephen Robert Frost).
  • May 2026: Held our second LNN lab day – project updates, a discussion of LLM usage in research, and a team-building event to close out the day.
  • May 2026: Welcomed Alice Schiavone to the LNN@UCPH, starting her PhD on the TRUSTMIND project – investigating how machine learning algorithms can be safely adapted and evaluated in clinical contexts, ensuring robustness and fairness by identifying different kinds of bias. Co-supervised with Martin Nørgaard, with collaboration from Irene (Zihui) Li's group at the University of Tokyo.
  • March 2026: Received DIKU's 2025 Dissemination Award, recognizing public engagement and science communication – including being selected by the Lundbeck Foundation to meet H.M. King Frederik X during the 2025 BrainPrize celebration.
  • March 2026: Invited keynote, "Efficient Data Management for Clinical Physicists", at the Nordic Association for Clinical Physics annual meeting in Göteborg, showcasing open-science tools for standardizing, processing and sharing clinical research data. Slides and outreach material on the OpenNeuroPET GitHub.
  • New book chapter, "The cause and effect of an MR image: Robustness and generalizability", published in Machine Learning in MRI, From Methods to Clinical Translation (Chapter 16), with Eike Petersen, Aasa Feragen and editor Thomas Küstner.

2025

  • The LNN lab handbook is live, produced during a lab hackathon.
  • The FAIMI 2025 online workshop (25 November) on Fairness of AI in Medical Imaging featured keynote speaker Leo Anthony Celi (MIT, MIMIC Project) alongside panelists Alceu Bisotto and Mauricio Reyes (Univ. of Bern), Elodie Germani (LTSI, France), and Ghada Zamzmi (HeartFlow). Recordings on the FAIMI YouTube channel.
  • The Lab for Neuroimaging and Neuroinformatics (LNN) @ UCPH officially launched, co-led with Martin Nørgaard, focused on advanced imaging technologies, neuroinformatics, machine learning and image analysis to understand brain structure, function and connectivity.
  • Former MSc student Hristo Georgiev presented work on generating synthetic head-motion curves for MRI at the 1st MICCAI Reconstruction and Imaging Motion Estimation (RIME) Workshop, 27 September.
  • Voted ISMRM Motion Correction study group secretary, beginning a three-year progression (secretary, vice-chair, chair, former chair) leading up to ISMRM 2028 in Copenhagen.
  • Abstract "Assessing the Influence of Preprocessing on the Agreement of Image Quality Metrics with Radiological Evaluation in the Presence of Motion" accepted at ISMRM 2025 as a power pitch (Hannah Eichhorn, Elisa Marchetto, Daniel Gallichan, Julia A. Schnabel).
  • Awarded a large Lundbeck Foundation grant for TRUSTMIND: AI in Mental Health – Ensuring Evaluation, Trustworthiness, Ethical and Legal Considerations, with Aasa Feragen, Merete Osler, Sune Holm and Katharina Ó Cathaoir. The 5-year project develops AI algorithms for screening Major Depressive Disorder, guided by ethical and legal principles. It grew out of the 2019–2020 "Bias and Fairness in Medicine" project (with Aasa Feragen and Sune Holm) and helped seed the FAIMI initiative. See the Lundbeckfonden announcement.
  • We co-organise a pre-ACL workshop in Copenhagen (26 July), supported by DDSA and the Pioneer Centre for AI.
  • Our group presented 3 papers at ACL 2025: two main-conference papers on fact checking and retrieval-augmented generation, and a CL survey on cross-cultural NLP.

2024

  • Second poster session at the ISMRM MoCo workshop 2024: Vincent Beliveau presented "Generative Models for Children’s Head Motion in Resting-State fMRI" (H. A. Georgiev, J. F. Sheye, V. Beliveau, M. Ganz).
  • First poster session at the ISMRM MoCo workshop 2024: posters on "Head Movement in the Resting-State fMRI Data of the ABCD Study" (J. F. Sheye, H. A. Georgiev, M. Ganz) and "Simulating Head Motion in MRI: A Silicone Phantom Approach with Machine Learning Integration" (Mads Daugaard, Emil Riis-Jacobsen, Hang Yin, Kenny Erleben, Melanie Ganz).
  • The MoCo project was represented heavily at the ISMRM MoCo workshop, Quebec (3–6 September): an invited talk, "Image Quality Metrics for Motion Assessment", plus three accepted abstracts.
  • Invited talk at the GRC In Vivo Magnetic Imaging Conference, in the session "AI in MRI: What About Generalizability, Bias and Fairness?", alongside Prof. Daniel Rueckert (TU Munich/Imperial College London) and Prof. Daniel Alexander (UCL).
  • Presented a poster at MIDL 2024 on "Classification of Medical Text in Small and Imbalanced Datasets in a Non-English Language", with postdoc Vincent Beliveau and colleagues from NRU (BrainDrugs, braindrugs.nru.dk).
  • Interviewed by TV2 news on AI algorithms in cancer diagnostics.
  • Congratulations to Madeleine Wyburd on receiving a DDSA postdoctoral fellowship; she joined DIKU as part of the NIBS-CP team to work on infant brain segmentation.
  • The MoCo project had two abstracts accepted at FLUX and FIT'NG 2024, Baltimore; research nurse Thurid Waagstein Madsen presented.
  • Invited speaker, ISMRM Workshop on Motion Correction in MR, Quebec City, on "Image Quality Assessments for MRI".
  • Invited speaker, Gordon Research Conference (GRC) In Vivo Magnetic Resonance, on "AI in MRI: What About Generalizability, Bias and Fairness?", discussing with Prof. Daniel Rueckert and Prof. Daniel Alexander.
  • New ISMRM and OHBM abstracts accepted: three ISMRM abstracts from the MoCo project (image quality metrics, hybrid motion correction), and OpenNeuroPET abstracts including "Quantifying the impact of preprocessing strategies on statistical analyses" (Brice Ozenne, Martin Nørgaard, Cyril Pernet, Melanie Ganz).

2023

  • On 29 September 2023, the last MoCo project patient was successfully scanned: 47 of 50 children (94%) referred for cerebral MRI in anesthesia were scanned without anesthesia. A follow-up "mini-MoCo" project continues with 20 more children, driven by radiographers without the mock scanner, to inform MRI design for the new Mary Elizabeth's Hospital (BørneRiget).
  • MoCo and NIBS-CP projects presented at FLUX and FIT'NG 2023: co-organised the symposium "Emerging Methods for MRI in Infants and Young Children" with Rhodri Cusack, Cameron Ellis, Brittany Howell and Aine Dineen (slides on OSF); also participated in FIT'NG 2023 and visited neuroimaging labs at Stanford (Kalanit Grill-Spector, Cameron Ellis).
  • Co-organised the symposium "Imaging the Developing Brain – where do we go from here?" at Rigshospitalet (24–25 August), with over 80 participants and speakers including Alfred Peter Born, Marisa N. Spann (Columbia), Stefan Skare (Karolinska), David Edwards (King's College London), Terry Jernigan (UCSD) and Brittany Howell (Virginia Tech).
  • Organised a BIDS derivative meeting in Copenhagen (20–22 June) with Cyril Pernet, restructuring the BIDS specification's definition of an atlas (see the atlas BEP, BEP038).
  • Co-organised the FLUX 2023 symposium "Emerging Methods for MRI in Infants and Young Children"; abstract "Reducing the Need for General Anesthesia in Children Undergoing Neuroimaging by Preparation and Motion Correction" selected as a poster.
  • New abstracts accepted: "Selective MRI reacquisition to extend the working range of retrospective motion correction" (with TracInnovations and Lars Hanson, DRCMR) as an ISMRM oral, and "A PET Molecular Imaging Brain Atlas of Cyclooxygenase-1" (with Douglas N. Greve, Martinos Center, and NIMH collaborators).
  • The MoCo project continued scanning participants through the year, on track to complete 50 children; first data release on OpenNeuro, preprint on PsyArXiv, and reproduction scripts on GitHub.

2022

  • First NIBS-CP site visit, with Kathrine Skak Madsen, to the Trinity College Institute for Neuroscience (Rhodri Cusack's lab on wake baby scans).
  • Results of the BIDS Steering Group election: succeeded by Cyril Pernet (OpenNeuroPET collaborator) and Yaroslav Halchenko (Dartmouth College).
  • Co-organised "Responsible Machine Learning in Healthcare" workshop (27–28 October) as part of the Bias and Fairness in Medicine project; recordings on YouTube.
  • NIBS-CP grant funded: Kathrine Skak Madsen (PI) and Melanie Ganz (Co-PI) received 10 mio DKK from the Elsass Foundation for "NeuroImaging of Babies during natural Sleep to assess typical development and Cerebral Palsy (NIBS-CP)", establishing procedures for infant/toddler MRI during natural sleep without sedation or GA in Denmark.
  • Co-organised the Fairness of AI in Medical Imaging (FAIMI 2022) workshop (5 October); recordings on YouTube.
  • Eike Petersen presented "Feature robustness and sex differences in medical imaging: a case study in MRI-based Alzheimer's disease detection" at MICCAI 2022, Singapore.
  • Panel speaker on data sharing at the first FITNG (Fetal, Infant, Toddler Neuroimaging Group) conference, Paris.
  • Presented work on characterising children's head motion at the ISMRM Workshop on Motion Detection & Correction, Merton College, Oxford.
  • Participated in the DIKU–DTU summer school 2022, "Human in the loop".
  • Lectured at NeuroHackademy 2022 (recordings online).
  • Presented "Evaluating the match of image quality metrics with radiological assessment in a dataset with and without motion artifacts" at ISMRM 2022, London: no single image-quality metric proved sensitive to all artifact types, though SSIM/PSNR (full-reference) and Image Entropy/Average Edge Strength/Tenengrad (reference-free) correlated best with radiological scores.

2021

  • Three abstracts presented at NRM2021: the PET-BIDS extension standard, estimating PET partial volume FWHM directly from data, and a high-resolution atlas of the brain's benzodiazepine binding site of GABAA receptors.
  • Hosted Brainhack Global Copenhagen.
  • Rigshospitalet's media team featured a project update, including the mock scanner.
  • Presented MoCo project results (healthy adult volunteers) and app/mock-scanner progress to the Elsass Foundation (24 September).
  • Started the prospective clinical study combining preparation and motion correction in pediatric MRI.
  • Hosted the RealNoiseMRI challenge at MICCAI 2021 (27 September), reopened for a second round at MedNeurIPS 2021.
  • Transitioned from the Machine Learning to the Image section, Department of Computer Science, as Associate Professor (1 September 2021).
  • Invited speaker at the EANM 2021 Pre-Congress Symposium "PET/MR – The Cross Path of Morphology and Functionality", on MRI motion correction.
  • Radiography students Martin and Bianca completed their project evaluating prospectively motion-corrected scans.
  • OHBM abstract, "Prospective motion correction and reacquisition in a complete clinical protocol for brain MR imaging".
  • Public-outreach talk for Forskningens Døgn / IDA (Danish engineering union) on scanning children without anesthesia.
  • The PET Brain Imaging Data Structure (BIDS) released with BIDS v1.6.0.
  • Article "A high-resolution in vivo atlas of the human brain's benzodiazepine binding site of GABAA receptors" published in NeuroImage.
  • OpenNeuroPET project kicked off with the PET BIDS community review, gathering final feedback on the PET BIDS standard (validator extension and BIDS examples).

2020

  • The Novo Nordisk Foundation funded the OpenNeuroPET research infrastructure.
  • The motion correction project was featured on the back cover of Rigshospitalet's IndenRigs magazine (December).
  • The Bias and Fairness project team published an article in Videnskab.dk's Forskerzone.
  • Congratulations to Marc Fuglsang Christensen and Anders Lykkebo-Valløe on completing Master's theses on VR-gamified MRI patience training and psilocybin brain states.
  • PET BIDS workshop highlights released ahead of OHBM, on YouTube.
  • Giske Opheim presented "Optimizing a preprocessing pipeline for structural 7T MR analyses in FreeSurfer" at ISMRM.
  • Congratulations to Ella Hedebø, Emil Rasmussen, Marcus Astrup Hansen and Andreea-Veronica Vascan on completing Bachelor's/Master's theses.
  • BIDS town hall and PET BIDS extension talks from the OHBM Open Science Room made available for re-watching.
  • Martin Nørgaard and Douglas N. Greve presented work at OHBM 2020 on the GABA-A atlas and PET partial volume estimation.
  • Presented the BIDS PET extension at the OHBM Open Science Room.
  • Opinion piece in the Danish newspaper Politiken, timed with the launch of "Bias and Fairness in Medicine".
  • "Bias and Fairness in Medicine" project started (1 April); project website at fairmed.compute.dtu.dk.
  • Motion correction work featured on Danish national evening news (DR) at the opening of Rigshospitalet's North Wing, attended by the Queen.

2019

  • Featured in short outreach videos made by Informationens medieskole about the children's MoCo project.
  • Hosted visitors Andre van der Kouwe and Stephen Robert Frost (Martinos Center) to set up prospective motion correction sequences on the new NRU research scanner.
  • Paper "The structure of the serotonin system: A PET imaging study" (with Vincent Beliveau) published in NeuroImage.
  • Elected member of the BIDS Steering Group, after being part of the BIDS community since fall 2017.
  • Received Danish Research Council funding for "Bias and Fairness in Medicine" with Aasa Feragen and Sune Holm.
  • Presented "Preprocessing, Prediction and Significance: Framework and Application to Brain Imaging" at MICCAI 2019, Shenzhen.
  • "Different Preprocessing Strategies Lead to Different Conclusions: A [11C]DASB-PET Reproducibility Study" accepted at JCBFM.
  • Poster at the 32nd ECNP Congress on preprocessing strategies in PET.
  • Presented "The Structure of the Serotonin System: A PET Imaging Study" at the BrainPET meeting, Yokohama.
  • "Optimization of Preprocessing Strategies in Positron Emission Tomography (PET) Neuroimaging: A [11C]DASB PET Study" accepted in NeuroImage.
  • Congratulations to Martin Nørgaard on defending his PhD thesis (3 May).
  • Joined the BrainDrugs initiative (from 1 July), a 40 million DKK Lundbeck Foundation thematic alliance led by Gitte Moos Knudsen studying drug response prediction in epilepsy and depression.

2018

  • Collaborative work "Comparing fully automated state-of-the-art cerebellum parcellation from magnetic resonance images" (from the MICCAI ENIGMA challenge) published in NeuroImage.
  • Work with NRU PhD student Cheng Teng Ip on EEG/ERP reliability published.
  • "Multi-view consensus CNN for 3D facial landmark placement" (with Rasmus R. Paulsen) presented at ACCV 2018.
  • On maternity leave, June 2018 – March 2019, following the birth of a child.
  • Received nearly 3.9 mio DKK from the Elsass Foundation (April) for the 3-year project "How can you avoid to put children with cerebral palsy in general anesthesia while they are in a brain scanner?", hosted at NRU from April 2019.
  • Presented "The Impact of Preprocessing Pipeline Choice in Univariate and Multivariate Analyses of PET Data" at the 8th International Workshop on Pattern Recognition in Neuroimaging, Singapore.
  • Presented "False Positive Rates in PET" at the Neuro Receptor Mapping meeting, London.
  • Published review "Cerebral serotonin transporter measurements with [11C]DASB: A review on acquisition and preprocessing across 21 PET centres" in JCBFM.
  • Congratulations to Vincent Beliveau on successfully defending his PhD thesis, "Functional and Molecular Imaging of the Serotonin System in the Human Brain".

2017

  • Journal paper "Brain Networks Implicated in Seasonal Affective Disorder: A Neuroimaging PET Study of the Serotonin Transporter" published in Frontiers in Neuroscience.
  • Presented "Brain Networks involved in Seasonal Affective Disorder: A Neuroimaging PET Study of 5-HTT Expression" at OHBM, Vancouver.
  • Presented "Permutation tests for classification: Revisited" at PRNI 2017.
  • From May 2017, split time between DIKU (newly appointed Assistant Professor) and the Neurobiology Research Unit at Rigshospitalet, dividing time at DIKU between teaching and establishing hospital–university research collaborations.
  • Published "Cerebellar heterogeneity and its impact on PET data quantification of 5-HT receptor radioligands" in JCBFM.