Published June 26, 2026 | Version v1

BREIT: Derived Data and Pretrained Models for Brain Stroke Reconstruction with Multi-Frequency 3D EIT

  • 1. ROR icon CY Cergy Paris Université
  • 2. ROR icon Equipes Traitement de l'Information et Systèmes
  • 3. ROR icon Analyse, Géométrie et Modélisation
  • 4. ROR icon Centre de Vision Numérique

Description

Data and model release for BREIT, a framework for brain stroke reconstruction using multi-frequency 3D Electrical Impedance Tomography. This record includes BREIT derived head/tissue geometry tetrahedral element-wise data for ISLES-derived ischemic stroke cases, IXI-derived healthy controls, synthetic physiological cases, and derived UCLH patient geometries, as well as pretrained DeepDbar and dFNOBAR/FNO-U-Net cascade checkpoints. The release supports regeneration of forward EIT simulations, D-bar caches, and reconstruction experiments.

Files

SHA256SUMS.txt

Files (12.6 GB)

Name Size
md5:0af51ba6bd436d191dc9b9c1cb1b59e9
8.0 GB Download
md5:837cc2423efdbc24d5fc2946617e544e
1.7 GB Download
md5:fce05c86d6b183707324fde8bd850b34
2.2 GB Download
md5:1fbee350e61995f3e304f95a6d97f76b
121.2 MB Download
md5:248e9b276923db703144f1a976d94ea7
649.1 MB Download
md5:6439bdf0422ec7fb9f85858ed0a49636
526 Bytes Preview Download

Additional details

Software