SeedSeg

Prostate Gold Seed Segmentation
More Apps GitHub

Drop NIfTI or DICOM files

Converting...

50%
Status Ready

About SeedSeg

SeedSeg

Prostate gold seed segmentation in the browser

SeedSeg is a browser-based tool for automatic segmentation of gold fiducial markers in T1-weighted prostate MRI.

It uses a consensus of four 3D UNet models to detect three gold seed markers used in prostate radiotherapy treatment planning.

All computation runs locally in your browser using ONNX Runtime Web. No data is uploaded to any server.

Version

SeedSeg

Privacy

Your data never leaves your device.

SeedSeg runs entirely in your web browser using WebAssembly and ONNX Runtime. All processing occurs locally on your machine.

What this means

  • No data is uploaded to any server or cloud service
  • No internet connection is required after the page loads
  • Sensitive patient data remains on your computer at all times
  • There is no need to anonymise or de-identify your data before use

How it works

SeedSeg uses ONNX Runtime Web to run deep learning models directly in the browser via WebAssembly. Your NIfTI or DICOM files are read into browser memory, processed locally, and results are displayed without any network transfer.

Third-party services

SeedSeg does not use analytics, tracking, cookies, or any external services. The only network requests are loading the application itself and downloading model files (which are cached locally).

Citations

SeedSeg Code: GitHub | Data: OSF

Stewart, A. W., Goodwin, J., Richardson, M., Robinson, S. D., O'Brien, K., Jin, J., ... & Bollmann, S. (2023). Deep-Learning-Enabled Differentiation between Intraprostatic Gold Fiducial Markers and Calcification in Quantitative Susceptibility Mapping. bioRxiv, 2023-10.

DOI: 10.1101/2023.10.26.564293

SeedSeg uses the following methods. Please cite the relevant papers if you use these in your research.

Bias Field Correction

Homogeneity Correction Code: MriResearchTools.jl | WASM: QSMbly

Eckstein, K., Trattnig, S., Robinson, S. D. (2019). A Simple Homogeneity Correction for Neuroimaging at 7T. Proc. ISMRM 27th Annual Meeting.

Reference Implementations

The following repositories were used as references:

MriResearchTools.jl - Bias field correction algorithm
QSMbly - WASM build of bias field correction
niern2niern - NIfTI JavaScript utilities

dicompare Report