| { |
| "_ilex": { |
| "architecture": "ilex.models.triad.swin_model.TriadSwinViT", |
| "constructor_kwargs": { |
| "input_channels": 1 |
| }, |
| "format": "ilex", |
| "framework_version": { |
| "equinox": "0.13.8", |
| "ilex": "0.0.0.dev0", |
| "jax": "0.10.0", |
| "jaxlib": "0.10.0", |
| "numpy": "2.4.4", |
| "safetensors": "0.7.0" |
| }, |
| "has_state": false, |
| "origin": "ilex-native" |
| }, |
| "authors": "Wang S., et al.", |
| "copyright": "Network architecture and pretrained weights: copyright (c) the Triad authors, released under the MIT License. JAX / Equinox port: copyright (c) the ilex authors, released under the Apache-2.0 / GPL-3.0 dual license used by ilex itself.", |
| "data_type": "nibabel", |
| "description": "Triad vision foundation model for 3D MRI, ported to JAX / Equinox from the upstream PyTorch release. Triad is an nnUNet PlainConvEncoder pretrained self-supervised on Triad-131K (131,170 3D MRI volumes spanning brain, breast, and prostate; T1/T2/FLAIR/DWI/fMRI/DCE) and serves as a transfer-learning backbone for downstream MRI segmentation, classification, and registration. The published checkpoints are encoder-only (the self-supervised decoder / mask token are stripped); this port exposes the pretrained encoder, whose multi-scale features are the transfer representation. Two backbone families are ported: the nnUNet PlainConvUNet encoder (TriadPlainConvUNet) and the 3D Swin Transformer encoder (TriadSwinViT, the Swin-B variant, via the shared nimox SwinViT primitive). Each is released under two self-supervised objectives -- masked autoencoding (MAE) and SimMIM -- as separate bundles (four in total).", |
| "equinox_version": "0.13.8", |
| "ilex_version": "0.0.0.dev0", |
| "image_classes": "Single-channel 3D MRI volume (contrast-general; pretrained across T1, T2, FLAIR, DWI, fMRI, DCE).", |
| "intended_use": "Research. A pretrained 3D-MRI encoder backbone for transfer learning; consumers attach a task-specific decoder / head and fine-tune. Inputs are single-channel 3D MRI volumes with each spatial dimension a multiple of 32.", |
| "jax_version": "0.10.0", |
| "label_classes": "N/A -- self-supervised backbone; no fixed label set. Output is the tuple of per-stage encoder feature maps.", |
| "network_data_format": { |
| "inputs": {}, |
| "outputs": {} |
| }, |
| "numpy_version": "2.4.4", |
| "pred_classes": "Multi-scale encoder skip features (6 stages, channels [32, 64, 128, 256, 320, 320]); the bottleneck is the deepest skip.", |
| "references": [ |
| "Wang S., et al. (2025). Triad: Vision Foundation Model for 3D Magnetic Resonance Imaging. arXiv:2502.14064. https://arxiv.org/abs/2502.14064", |
| "Codebase: https://github.com/wangshansong1/Triad" |
| ], |
| "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json", |
| "task": "3D-MRI self-supervised foundation backbone (transfer learning)", |
| "version": "0.0.0" |
| } |
