| Apr 06, 2026 | Chiara’s work on training vessel segmentation models for diverse image modalities and spatial scales is now available in preprint: “VesSynth: Tubes Are All You Need for Robust Cross-Scale Cross-Modal 3D Vessel Segmentation”. I was also involved in three studies on HiP-CT imaging as part of the LINC consortium: |
| Mar 30, 2026 | JD’s and Gareth’s use of our “brain shape model” to test the validity of different MEG source reconstruction alorithms was published in Imaging Neuroscience: “Distorting anatomy to test MEG models and metrics”. |
| Nov 05, 2025 | I had the honor to contribute to Eugenio’s masterpiece “A probabilistic histological atlas of the human brain for MRI segmentation”, which introduces the NextBrain atlas and was published in Nature. Note that it now comes with a faster segmentation algorithm, lead by Oula and available in preprint: “Fast segmentation with the NextBrain histological atlas” |
| Jul 16, 2025 | Xiaoling’s paper “Learn2Synth: Learning Optimal Data Synthesis using Hypergradients for Brain Image Segmentation” was accepted at ICCV 2025 (Oct 19 – 23th, 2025)! |
| Jul 15, 2025 | Yu is presenting his paper “Schwarz–Schur Involution: Lightspeed Differentiable Sparse Linear Solvers” at ICML 2025 in Vancouver on Tuesday, 15 July. |
| May 13, 2025 | Chiara is presenting her work “A method for automatic 3D vasculature segmentation in ex vivo MRI using synthetic data”, for which she received a Magna Cum Laude award, in the Data Post-Processing oral session at ISMRM 2025 on Wednesday, 14 May. |
| Apr 03, 2025 | I will be presenting “A scalable computational framework for large-scale imaging of neural circuits” in the software demonstration session at OHBM 2025. I will also be presenting a poster on “SPM-python: Python Bindings for SPM25”, a project led by Johan Medrano, and will participate to the “30 years of SPM” roundtable. |
| Jan 21, 2025 | A new major version of the Statistical Parametric Mapping, SPM25, was released. This is the first major version of SPM since SPM12 (which, counter-intuitively, was released in 2014). It comes with a preprint and includes, among other novelties, the Multi-Brain (MB) and Factorisation-Based Image Labelling (FIL) toolboxes. |
| Nov 29, 2024 | In a new preprint, Xiaoling shows how to learn a synthesis strategy that is optimal for a given task: “Learn2Synth: Learning Optimal Data Synthesis Using Hypergradients”. |
| Jul 03, 2024 | Check out Etienne’s new preprint “Neurovascular Segmentation in sOCT with Deep Learning and Synthetic Training Data”! He is presenting a condensed version of this work at MIDL 2024 today. |
| Apr 27, 2024 | Etienne’s short paper on “label-free and data-free vasculature segmentation” was accepted at MIDL 2024! |
| Mar 01, 2024 | I am starting a new adventure at University College London as a Newton International Fellow, where I will work on High-resolution spatiotemporal models of the brain across the lifespan for diagnosis and decision-making. |
| Feb 26, 2024 | Sean’s paper on “Fully Convolutional Slice-to-Volume Reconstruction for Single-Stack MRI” was accepted at CVPR 2024! |
| Jan 16, 2024 | We have a website! |