Spinal cord MRI segmentation in the browser
Spinal Cord Toolbox is a browser-based tool for SCT stable spinal cord MRI segmentation workflows.
Segmentation runs locally using ONNX Runtime Web. Morphometry and lesion analysis run SCT Python in your browser, or the pinned SCT CLI on your chosen compute server. Uploaded masks and generated masks can be analyzed without an anatomy image.
Long runs can continue if this site is allowed to stay active in your browser.
Segmentation and browser SCT analysis keep your images and masks on this device. Selecting compute-server execution sends chosen masks only after you press Run analysis.
Spinal Cord Toolbox uses ONNX Runtime Web to run browser-compatible deep learning models directly in the browser via WebAssembly. Local segmentation reads NIfTI or DICOM images into browser memory. Browser SCT analysis runs Python in an isolated WebAssembly worker with local masks and exports. Native analysis uploads the selected NIfTI masks unchanged, runs the pinned SCT container, and downloads its original CSV or XLSX, pickle and labeled NIfTI artifacts.
The Neurodesk hosting layer records page views only, unless your browser sends Do Not Track or Global Privacy Control. It does not send custom events, patient image data, DICOM metadata, filenames, voxel values, masks, segmentations, screenshots, free-text patient-derived logs, or downloaded outputs to Google Analytics.
Browser SCT analysis downloads the pinned Pyodide Python runtime and scientific libraries from jsDelivr (cdn.jsdelivr.net) when it runs. These requests fetch fixed files and carry no images, masks, filenames or results.
De Leener B, Levy S, Dupont SM, Fonov VS, Stikov N, Louis Collins D, Callot V, Cohen-Adad J. SCT: Spinal Cord Toolbox, an open-source software for processing spinal cord MRI data. NeuroImage. 2017.
https://spinalcordtoolbox.com/stable/Spinal Cord Toolbox uses the following methods and tools:
Paszke A, et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library. NeurIPS 2019.
https://pytorch.orgMicrosoft. ONNX Runtime: cross-platform, high performance ML inferencing and training accelerator.
https://onnxruntime.aiLi X, Morgan PS, Ashburner J, Smith J, Rorden C. The first step for neuroimaging data analysis: DICOM to NIfTI conversion. J Neurosci Methods. 2016;264:47-56.
https://github.com/rordenlab/dcm2niixNiiVue Contributors. NiiVue: a WebGL2 medical image viewer.
https://github.com/niivue/niivue