My projects

Platforms & research that travel beyond the lab

Open-source infrastructure and medical imaging AI — from collaborative Kubernetes platforms to portable segmentation bundles and federated learning.

Platform

MAIA — Medical Artificial Intelligence Assistant

Open-source Kubernetes platform for collaborative medical AI. Integrates MONAI, 3D Slicer, OHIF, Orthanc, XNAT, and MLFlow into a cohesive environment for annotation, training, active learning, and clinical deployment. Published in npj Artificial Intelligence (2025).

  • Kubernetes
  • MONAI
  • MLOps
  • Clinical deployment
Overview NPJ article GitHub Docs
MAIA workspace

Bundle & federated learning

MONet Bundle

nnU-Net packaged as a MONAI Bundle for portable training, federated learning, and clinical deployment. Achieved 1st prize in BraTS 2025 Task 07 (GoAT) and powers multi-center lymphoma segmentation work presented at MICCAI DeCaF 2025.

  • nnU-Net
  • MONAI Bundle
  • NVIDIA FLARE
  • BraTS 2025
GitHub MONet-FL paper

Imaging research

Lung vessel connectivity & radio-dynamics

Anatomical priors for lung lobe segmentation (LVC maps in nnU-Net) and radio-dynamics features for pCR prediction in breast DCE-MRI — work aimed at more clinically reliable imaging biomarkers.

  • Deep learning
  • CT / MRI
  • Radiomics
JMI paper Hive_ML

Also contributing to

MONAI and nnU-Net — open tools that underwrite much of this stack.

Stack

Tools these projects run on

Containerized platforms, CI/CD, and medical imaging tooling shared across research and hospital environments.

  • Kubernetes / OpenShift
  • Docker / Helm
  • CI/CD · GitHub Actions · ArgoCD
  • MLOps · MLFlow · KubeFlow
  • Python · Bash
  • PyTorch · MONAI · nnU-Net
  • DICOM · PACS · Orthanc · OHIF
  • GCP · MinIO · Terraform · Ansible

Project publications

  • 2025
    MAIA: A Collaborative Medical AI Platform for Integrated Healthcare Innovation — npj Artificial Intelligence. Article · arXiv
  • 2025
    MONet-FL: Extending nnU-Net with MONAI for Clinical Federated Learning — LNCS, Springer / MICCAI DeCaF.
  • 2024
    Lung vessel connectivity map as anatomical prior knowledge for deep learning-based lung lobe segmentation — Journal of Medical Imaging. DOI

Full list on Home · Code on GitHub

Open to collaboration

If you are building medical AI platforms, deploying models in clinical research settings, or looking to run MAIA at your institution — I would like to hear from you.

Email Simone Explore MAIA Read notes