Initial release: BrainIAC ViT-B backbone + 4 downstream task heads (safetensors)

LICENSE

@@ -0,0 +1,33 @@
+BrainIAC Research-Only License
+
+Copyright © 2026
+Mass General Brigham and contributors
+
+Permission is hereby granted to use, copy, modify, and distribute this
+software and associated documentation files (the “Software”) solely for
+non-commercial academic research and educational purposes.
+
+Commercial use of the Software is strictly prohibited without a separate
+written license from Mass General Brigham. Commercial use includes, but is
+not limited to:
+- Use in products, services, or platforms offered for a fee or other
+  commercial advantage;
+- Use in clinical workflows, decision support systems, or healthcare
+  operations;
+- Use by for-profit entities for internal or external commercial purposes;
+- Redistribution or deployment of the Software or derived works as part
+  of a commercial offering.
+
+The Software may not be sublicensed, sold, leased, or otherwise transferred
+for commercial purposes without prior written authorization from
+Mass General Brigham.
+
+The Software is provided “AS IS”, without warranty of any kind, express or
+implied, including but not limited to the warranties of merchantability,
+fitness for a particular purpose, and non-infringement. In no event shall
+the authors or copyright holders be liable for any claim, damages, or other
+liability, whether in an action of contract, tort, or otherwise, arising
+from, out of, or in connection with the Software or the use or other
+dealings in the Software.
+
+All commercial rights are expressly reserved.

README.md

@@ -0,0 +1,118 @@
+---
+license: other
+license_name: mass-general-brigham-non-commercial
+license_link: LICENSE
+tags:
+  - brain
+  - mri
+  - neuroimaging
+  - vit
+  - foundation-model
+  - medical-imaging
+library_name: brainiac-rs
+pipeline_tag: feature-extraction
+---
+
+# BrainIAC — Brain Imaging Adaptive Core
+
+**A generalizable foundation model for analysis of human brain MRI**
+
+BrainIAC is a Vision Transformer (ViT-B/16) pretrained with SimCLR on structural brain MRI scans.
+Published in [Nature Neuroscience](https://www.nature.com/articles/s41593-026-02202-6) (2026).
+
+## Model Details
+
+| Property | Value |
+|----------|-------|
+| Architecture | MONAI ViT-B/16³ (3D) |
+| Parameters | 88.4M |
+| Input | 96×96×96 single-channel brain MRI |
+| Patches | 216 (6×6×6 grid, 16³ voxel patches) |
+| Hidden dim | 768 |
+| Layers | 12 transformer blocks |
+| Heads | 12 attention heads |
+| MLP dim | 3072 |
+| Pretraining | SimCLR contrastive learning |
+| Output | 768-dim feature vector (first patch token) |
+
+## Files
+
+- `backbone.safetensors` — Pretrained ViT backbone weights
+- `config.json` — Model configuration
+- `LICENSE` — Non-commercial academic research license
+
+## Downstream Tasks
+
+The backbone can be fine-tuned for:
+- **Brain age prediction** (regression)
+- **IDH mutation classification** (binary, dual-scan FLAIR+T1CE)
+- **MCI classification** (binary)
+- **Glioma overall survival** (binary, quad-scan T1+T1CE+T2+FLAIR)
+- **MR sequence classification** (4-class: T1/T2/FLAIR/T1CE)
+- **Time-to-stroke prediction** (regression)
+- **Tumor segmentation** (UNETR decoder)
+
+## Usage with brainiac-rs (Rust)
+
+```bash
+cargo run --release --bin infer -- \
+    --weights backbone.safetensors \
+    --input brain_t1.nii.gz
+```
+
+```rust
+use brainiac_rs::{BrainiacEncoder, TaskType};
+
+let (encoder, _) = BrainiacEncoder::<B>::load(
+    "backbone.safetensors", None,
+    TaskType::FeatureExtraction, 1, device,
+)?;
+let features = encoder.encode_nifti(Path::new("brain.nii.gz"))?;
+// features: Vec<f32> with 768 dimensions
+```
+
+## Usage with Python
+
+```python
+import torch
+from monai.networks.nets import ViT
+from safetensors.torch import load_file
+
+model = ViT(in_channels=1, img_size=(96,96,96), patch_size=(16,16,16),
+            hidden_size=768, mlp_dim=3072, num_layers=12, num_heads=12)
+
+weights = load_file("backbone.safetensors")
+model.load_state_dict(weights, strict=False)
+model.eval()
+
+# features[0][:, 0] gives the 768-dim feature vector
+features = model(preprocessed_mri)
+```
+
+## Preprocessing
+
+Input MRI volumes must be:
+1. Skull-stripped (HD-BET recommended)
+2. Registered to standard space (MNI152)
+3. Bias field corrected (N4)
+4. Resized to 96×96×96 voxels (trilinear)
+5. Z-score normalized (nonzero voxels only)
+
+## Citation
+
+```bibtex
+@article{tak2026generalizable,
+  title={A generalizable foundation model for analysis of human brain MRI},
+  author={Tak, Divyanshu and Gormosa, B.A. and Zapaishchykova, A. and others},
+  journal={Nature Neuroscience},
+  year={2026},
+  publisher={Springer Nature},
+  doi={10.1038/s41593-026-02202-6}
+}
+```
+
+## License
+
+This model is licensed for **non-commercial academic research use only**.
+Commercial use requires a separate license from Mass General Brigham.
+See [LICENSE](LICENSE) for details.

backbone.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:12b7745715724862c92e7b0bb1252d257abf8a7af055122cb09771368831e275
+size 353376032

brainage_head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:920fd51409f9ebdfe3f0a53d5eea5836803d4446b2e96e0a93f404890c419797
+size 3220

config.json

@@ -0,0 +1,16 @@
+{
+  "model_type": "brainiac-vit",
+  "architecture": "MONAI ViT-B/16\u00b3",
+  "img_size": 96,
+  "patch_size": 16,
+  "in_channels": 1,
+  "hidden_size": 768,
+  "mlp_dim": 3072,
+  "num_layers": 12,
+  "num_heads": 12,
+  "num_patches": 216,
+  "norm_eps": 1e-06,
+  "pretraining": "SimCLR",
+  "input_format": "NIfTI (.nii.gz), skull-stripped, registered, 96\u00d796\u00d796",
+  "preprocessing": "trilinear resize to 96\u00b3 + z-score normalization (nonzero voxels)"
+}

idh_head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:257fdbabfe6d09dc1d99dd59c0db648ce1a9526df1680764d6c25181b0712a01
+size 3220

mci_head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c859063b6529f370ea198aede2f5a92998b9055d8db185c86c8a74c8a74cc27
+size 3220

stroke_head.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b51e35274929ce06f2feb46089c70689d0952dc4d22fa8125000fda98bd8bb6a
+size 3220