SpatioTemporal Labeler is a cross-platform desktop editor for manual and semi-automatic 3D/4D medical image segmentation. It supports dynamic and time-resolved MRI, including 4D flow MRI, in NIfTI and NRRD files, with linked spatial-temporal views, real-time 3D label rendering, temporal propagation, and metadata-preserving I/O.
- Linked X-Y/X-Z/Y-Z spatial and X-T/Y-T/Z-T temporal views with a shared X/Y/Z/T cursor
- Peak-signal frame selection, time navigation, and display-only temporal-axis stretch
- Current-frame or all-frame label editing with physical brush/eraser footprints and one-gesture undo
- Adjacent-frame temporal propagation using physical displacement, intensity matching, and label/threshold barriers
- Synchronized 2D overlays and independently colored 3D label surfaces while navigating time
- Separate threshold-mask constraints with all-frame editing and threshold-bypass modes
- Multiple image/label sequences with RAS-mapped previews and 3D-label-to-4D frame mapping
- Metadata-preserving 3D/4D NRRD and NIfTI read/write
Download the package for your platform from the latest release. Portable packages include Python, Qt, VTK, and all runtime dependencies.
| Platform | Release asset | Run |
|---|---|---|
| Windows 10/11 x64 | SpatioTemporalLabeler-<version>-windows-x64.zip |
Extract and open SpatioTemporalLabeler/SpatioTemporalLabeler.exe |
| Linux x86_64 | SpatioTemporalLabeler-<version>-linux-x86_64.tar.gz |
Extract and run SpatioTemporalLabeler/SpatioTemporalLabeler |
Each package also contains per-user install and uninstall scripts. No administrator access is required.
Every release includes a pure Python wheel and source distribution. Install the current release directly from GitHub:
python -m pip install "https://github.com/AssociatedPrimeIdeal/SpatioTemporalLabeler/releases/download/v0.4.0/spatiotemporal_labeler-0.4.0-py3-none-any.whl"Alternatively, download the wheel from the release and install it locally:
python -m pip install spatiotemporal_labeler-0.4.0-py3-none-any.whlLaunch the installed application with spatiotemporal-labeler. Python 3.9 or newer is required. Runtime dependencies are installed automatically by pip.
The repository includes an 18-frame PCMRA image and matching label sequence in examples/sample-data.
spatiotemporal-labeler examples/sample-dataWhen a directory is provided, every direct .nrrd, .nii, and .nii.gz file is loaded. Files whose names contain seg, mask, or label are opened as label sequences; the remaining files are opened as image sequences.
When present, pcmra.seq.nrrd is selected as the initial display image and seg.seq.nrrd as the initial label sequence. A 4D image initially opens on the earliest frame with the largest whole-frame sum of absolute finite voxel intensities; NaN and infinite values are ignored.
You can also launch without arguments and load or drop NRRD/NIfTI files:
spatiotemporal-labeler| Input | Action |
|---|---|
B, E, S, L, G |
Brush, eraser, scissors lasso, contour fill, or temporal propagation |
Hold I / H |
Pick labels / hide 2D label overlays |
Press/release CapsLock, Q, or W |
Toggle all-frame threshold-constrained, current-frame threshold-bypass, or all-frame threshold-bypass editing; hold during a stroke for a temporary mode |
[ / ] |
Decrease/increase brush or eraser diameter |
Left / Right |
Previous/next time frame |
Up / Down |
Next/previous orthogonal spatial slice |
V |
Cycle X-T/Y-T/Z-T display-only time-axis stretch (1x/2x/4x) |
R |
Reset the hovered preview or the main 2D views |
Enter / Esc |
Apply a pending contour / cancel a contour or lasso |
Ctrl+S / Ctrl+Shift+S |
Save / save as |
Ctrl+W |
Close all loaded files |
Ctrl+Z / Ctrl+Y / Ctrl+Shift+Z |
Undo / redo |
| Input | Action |
|---|---|
| Left drag | Use the selected editing tool in a spatial view |
| Right drag | Temporarily erase without changing the selected tool |
Hold Shift and move / Shift + left drag |
Locate the shared cursor / pan a 2D view |
| Middle drag in 2D / middle or right drag in 3D | Adjust window width and window level / pan and zoom |
Ctrl + wheel / Shift + wheel |
Zoom a 2D view / change brush diameter |
| Wheel in a spatial view | Change its orthogonal slice |
| Alt + left drag in 3D | Rotate the 3D camera |
| Double-click | Activate an image preview, change a label color, maximize a view, or confirm a contour depending on the target |
| Drag a locator-line arrow | Move the linked X, Y, or Z cursor coordinate |
Enable All time frames to apply ordinary spatial gestures in every frame. Temporal propagation has its own default all-frame range and does not use this option. It begins from a spatial-view source patch, then independently matches each voxel only into adjacent frames; it does not grow within a frame and stops in a direction after an unmatched frame.
Image and label sequences are normalized internally to canonical RAS [X,Y,Z,T]; 3D sources use a singleton T axis. Saving reverses the source transform and preserves the original dimensionality and relevant NRRD/NIfTI metadata. When a spatially matching 3D label sequence is opened over a 4D image, it can be copied to every frame (the default) or placed in one selected frame; the mapped result becomes a new unsaved 4D label sequence. Other editing requires a matching voxel grid.
SpatioTemporal Labeler is distributed under the GNU General Public License v3.0.