Skip to content

code.markdown

Latest commit

 

History

History
170 lines (154 loc) · 9.16 KB

code.markdown

File metadata and controls

170 lines (154 loc) · 9.16 KB
layout page
title Code & Data
permalink /code/
lede Open-source code and datasets from MINNE Lab. We aim to publish data and code from our projects whenever possible.

Source code

GitHub organisation

minnelab

The lab's GitHub organisation. Research platforms, publication code, teaching material, and federated-learning infrastructure live here.

Open on GitHub →

Research platforms

  • Medical AI platform

    MAIA

    Kubernetes-based platform for collaborative medical-AI research — data preparation, training, active learning, deployment, and evaluation in one environment.

  • Federated learning

    MONet-Bundle

    MONAI bundle integrating nnU-Net for clinical federated learning, active learning, and PACS integration (MICCAI 2025 DeCaF).

  • Infrastructure

    NVFlare-Dashboard

    NVIDIA FLARE dashboard for provisioning and monitoring federated-learning experiments across institutions.

  • Detection framework

    nnDetection

    Self-configuring framework for 3D medical-object detection. Our fork with lab-specific integrations.

Code accompanying publications

  • Scientific Reports · 2025

    DBM_with_DL

    Decomposing the effect of normal aging and Alzheimer's disease in brain morphological changes via learned aging templates.

  • Medical Image Analysis · 2025

    InBrainSyn

    Synthesising individualised aging brains in health and disease with generative models and parallel transport.

  • ISBI · 2025

    longitudinal-rigid-registration

    Learning accurate rigid registration for longitudinal brain MRI from synthetic data.

  • Tractography

    rCOMMIT

    Randomly COMMITting — iterative convex optimisation for microstructure-informed tractography.

  • ADSMI @ MICCAI · 2024

    DA_nnUNet

    Unsupervised domain adaptation for pediatric brain-tumor segmentation.

  • HBM · 2023

    Synthetic-Brain-Aging

    Fast 3D image generation for healthy brain aging using diffeomorphic registration.

  • Breast ultrasound

    DynamicBUS

    Diffusion-model-based breast ultrasound video generation for diagnosis.

Teaching & workshops

For a full list — including internal tools and configuration repositories — see the minnelab organisation on GitHub.

Rodrigo Moreno — personal repositories

GitHub user

rodrigomorenokth

Earlier reference implementations published directly under Rodrigo Moreno's personal GitHub account.

Open on GitHub →
  • Vascular imaging

    RPD

    Ring Pattern Detector — reference implementation of the vesselness filter for tubular-structure extraction.

  • Diffusion MRI

    GMIL

    Generalised Mean Intercept Length Tensor — anisotropy estimation for orientation-distribution data.

Datasets

Open dataset

Synthetic Healthy Brain Aging MRIs with Segmentation Masks

A collection of 6,448 synthetic T1-weighted MRI scans simulating longitudinal brain aging, each with a corresponding segmentation mask. Generated using deep generative models to support research in neuroimaging and healthy aging.

Modality
T1-weighted MRI (synthetic)
Resolution
160 × 160 × 192 voxels, 1 × 1 × 1 mm
Volume
6,448 scans, 52.4 GB
Citation

Jingru Fu, Antonios Tzortzakakis, José Barroso, Eric Westman, Daniel Ferreira, Rodrigo Moreno (2023). Synthetic healthy brain aging MRIs with segmentation masks. doi:10.23698/aida/synthetic/shbamri

Related publication: Fu, J. et al. Fast 3D image generation for healthy brain aging using diffeomorphic registration. doi:10.1002/hbm.26165