fidoptGaussian blur
SimpleITK's recursive Gaussian smoothing at its default standard deviation, 1 mm along every axis: the simplest denoiser, no learnt weights.
SynthesisCPUversion SimpleITK 2.5.6MR, MRApache-2.0
Run it
With its KonfAI app, at the benchmark's settings (the fidopt app repositories are private for now: a Hugging Face token with access opens them):
pip install konfai-apps
| setting | Command | |
|---|---|---|
| Gaussian blur | Get the app | konfai-apps infer fideus-labs/fidopt-synthesis:Gaussian_Blur -i image.nii.gz -o out |
| sigma 0.55 mm | Get the app | konfai-apps infer fideus-labs/fidopt-synthesis:Gaussian_Blur -i image.nii.gz -o out --set=sigma=0.55 |
Where it works
Its place on each kind of problem it has results on, among the methods ranked there. A method that leads one problem can trail another: read the one closest to yours.
| Problem | setting | Place | Ranked | Datasets |
|---|---|---|---|---|
| MR to MR, one acquisition to the mean of its repetitions, synthesis | 3 | of 3 | 1 of 1 |
Results
| Dataset | setting | # | Read by | Score | Failed | Time s |
|---|---|---|---|---|---|---|
| M4Raw, one 0.3 T T2 repetition to the mean of the others | 3 | MAE | 2.9 | 0 | 5 | |
| M4Raw, one 0.3 T T2 repetition to the mean of the others | sigma 0.55 mm | MAE | 2.5 | 0 | 5 |
- Reference
- Deriche, INRIA RR-1893 1993; Lowekamp et al., Front. Neuroinform. 2013