ilex-hub/synthmorph.deform.3

SynthMorph deformable (HyperMorph) -- Deformable variant v3 (FreeSurfer 8.x default)

Description

SynthMorph deformable (Hoffmann et al. 2024, Imaging Neuroscience 2:1-33) ported to JAX / Equinox from the FreeSurfer-bundled VoxelMorph reference implementation. The network is a HyperMorph-style hypernetwork wrapped around a 5-level VxmDense U-Net -- a 4-layer dense MLP (32 units per layer) takes a single hyperparameter (the deformation regularization weight) and outputs the conv kernels + biases for all 13 conv layers of the embedded U-Net at forward time. The published synthmorph.deform.3.h5 (~3.5 GB) carries the dense-projection weights that map the hypernet embedding to each conv layer's kernel + bias; the conv layers themselves hold no static weights. The v0 ilex bundle returns the raw 3-channel velocity field at the input spatial resolution; downstream integration (squaring-and-scaling VecInt) and spatial-transform warp are parameter-free pure numerics and live outside the v0 bundle, mirroring the affine port's barycenter / fit_affine deferral.

Intended use

Hypernetwork-conditioned deformable registration of two 3D brain volumes. The user picks a regularization weight in [0, 1] at inference time without retraining. The bundle returns the raw 3-channel velocity field; downstream integration (squaring-and-scaling VecInt) and spatial-transform warp are parameter-free and live outside the v0 bundle.

Usage

from ilex.models.synthmorph_deform import SynthMorphDeform
model = SynthMorphDeform.from_pretrained('ilex-hub/synthmorph.deform.3')

Authors

Hoffmann M., Hoopes A., Greve D. N., Iglesias J. E., Fischl B., Dalca A. V.

Citation

Hoffmann M., Hoopes A., Greve D. N., Fischl B., Dalca A. V. (2024). Anatomy-aware and acquisition-agnostic joint registration with SynthMorph. Imaging Neuroscience, 2:1-33. doi:10.1162/imag_a_00197. Original HyperMorph framework: Hoopes A., Hoffmann M., Greve D. N., Fischl B., Guttag J., Dalca A. V. (2022). Learning the effect of registration hyperparameters with HyperMorph. Journal of Machine Learning for Biomedical Imaging, 1:1-30.

References

• Hoffmann M., Billot B., Greve D. N., Iglesias J. E., Fischl B., Dalca A. V. (2022). SynthMorph: learning contrast-invariant registration without acquired images. IEEE Transactions on Medical Imaging, 41(3):543-558. doi:10.1109/TMI.2021.3116879.
• Hoffmann M., Hoopes A., Greve D. N., Fischl B., Dalca A. V. (2024). Anatomy-aware and acquisition-agnostic joint registration with SynthMorph. Imaging Neuroscience, 2:1-33. doi:10.1162/imag_a_00197.
• Hoopes A., Hoffmann M., Greve D. N., Fischl B., Guttag J., Dalca A. V. (2022). Learning the effect of registration hyperparameters with HyperMorph. Journal of Machine Learning for Biomedical Imaging, 1:1-30.

License

HF Hub license tag: other HF Hub license slug: freesurfer-research

Effective terms: Weights distributed by upstream as part of the FreeSurfer software bundle under the FreeSurfer Software License (FSLA), a permissive academic / non-commercial research offering. See license_url for the binding terms. Same license terms as the synthmorph_affine catalog entry; the deformable hypernetwork is a separately-trained downstream artefact of voxelmorph + SynthMorph synthetic-data training.

Upstream license reference: https://surfer.nmr.mgh.harvard.edu/fswiki/FreeSurferSoftwareLicense

Copyright

Network architecture, training code, and pretrained weights: copyright (c) the SynthMorph / VoxelMorph authors and the FreeSurfer maintainers, distributed via the FreeSurfer software distribution under the FreeSurfer Software License (FSLA; permissive academic / non-commercial research use). See https://surfer.nmr.mgh.harvard.edu/fswiki/FreeSurferSoftwareLicense for the binding terms. The voxelmorph reference implementation itself is dual- Apache-2.0 / GPL-3.0; the SynthMorph weights are a downstream artefact of voxelmorph + synthetic-data training, distributed through the FreeSurfer bundle. JAX / Equinox port code: copyright (c) the ilex authors, released under the Apache-2.0 / GPL-3.0 dual license used by ilex itself; the ilex port covers only the original Equinox re-expression and does not override the upstream FreeSurfer / voxelmorph terms.

Upstream source

Original weights / reference implementation: https://github.com/voxelmorph/voxelmorph

Provenance

This artefact was produced by ilex's save/load pipeline. The architecture is implemented in ilex.models.synthmorph_deform.SynthMorphDeform and the weights have been converted from their upstream format. See the upstream source above for the canonical reference.