scalesufer

scalesurfer is a repository for fast FreeSurfer inference. The volumetric model uses UNet with a Transformer bottleneck. The stats models predict FreeSurfer measures, including cortical thickness, surface area, curvature, and folding index.

Installation

pip install scalesurfer

Usage

See the inference notebook for additional settings for faster processing.

from scalesurfer import ScaleSurfer

# Anatomical images
adni_dir = "/home/rph/scalesurfer/data/adni_bids"
subjects = ["sub-002S0559", "sub-002S0619"]

anat_files = [
    f"{adni_dir}/sub-002S0559/ses-20060627/anat/sub-002S0559_ses-20060627_T1w.nii.gz",
    f"{adni_dir}/sub-002S0619/ses-20060601/anat/sub-002S0619_ses-20060601_T1w.nii.gz"
]

# Predict aparc+aseg and stats tables
surfer = ScaleSurfer(anat_files, subjects, "/tmp/scalesurfer_subjects", device="cuda")
surfer.prepare_images()
surfer.predict_volumes()
surfer.plot_volume(subjects[0])
df_stats = surfer.predict_stats()

GPU

All models are implemented with pytorch and inference time depends on GPU. CPU-based inference will be much slower. These models were developed on an NVIDIA card with 32 GB of VRAM and 64 RAM. With this hardware, the settings in the inference notebook worked well. Please open an issue if inference fails on your hardware.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

scalesurfer-0.1.2.tar.gz (140.6 kB view details)

Uploaded Jul 14, 2026 Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

scalesurfer-0.1.2-py3-none-any.whl (159.5 kB view details)

Uploaded Jul 14, 2026 Python 3

File details

Details for the file scalesurfer-0.1.2.tar.gz.

File metadata

Download URL: scalesurfer-0.1.2.tar.gz
Upload date: Jul 14, 2026
Size: 140.6 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for scalesurfer-0.1.2.tar.gz
Algorithm	Hash digest
SHA256	`947e895c5536b8d05bf7c1917c5d3cfc546dab3945e26bf5a9dc2f0deae63233`
MD5	`869209e5dc5c9ada2c8c8489d4b9e4a1`
BLAKE2b-256	`62b4aa8dea773c3df2beb2f4fd0ba7b739d02f833b1504bfbe57719c96271c95`

See more details on using hashes here.

Provenance

The following attestation bundles were made for scalesurfer-0.1.2.tar.gz:

Publisher: publish.yml on voytekresearch/scalesurfer

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: scalesurfer-0.1.2.tar.gz
- Subject digest: 947e895c5536b8d05bf7c1917c5d3cfc546dab3945e26bf5a9dc2f0deae63233
- Sigstore transparency entry: 2170901084
- Sigstore integration time: Jul 14, 2026
Source repository:
- Permalink: voytekresearch/scalesurfer@71049255f2328763c77219bb0b813391f214cd7c
- Branch / Tag: refs/heads/master
- Owner: https://github.com/voytekresearch
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@71049255f2328763c77219bb0b813391f214cd7c
- Trigger Event: workflow_dispatch