fidoptmultiGradICON
unigradicon-register --model multigradicon with its defaults: the network's prediction, then 50 steps of instance optimisation on LNCC.
RegistrationGPUversion unigradicon 1.0.5, weights 1.0CT, CBCT, 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 konfai-apps infer fideus-labs/fidopt-registration:MultiGradICON -i fixed.nii.gz -i moving.nii.gz -o out --gpu 0
The app on Hugging Face: its model, its settings and how it is scored.
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 | Place | Ranked | Datasets |
|---|---|---|---|
| Temporal bone CBCT, between donors | 3 | of 4 | 1 of 1 |
| Pediatric T1 MRI with automatic regional labels | 4 | of 4 | 1 of 1 |
| Abdomen MR, inter-patient | not ranked yet | 1 of 2 |
Results
| Dataset | # | Read by | Score | Change | SDlogJ | Failed | Time s |
|---|---|---|---|---|---|---|---|
| ACDC | 1 | Dice | 0.817 | +0.259 | 0.031 | 0 | 76 |
| Pediatric T1 MRI with and without previous ADHD diagnosis | 4 | Dice | 0.588 | +0.176 | 0.228 | 0 | 97 |
| AMOS MR/MR | 3 | Dice | 0.410 | +0.145 | 0.161 | 0 | 93 |
| OpenEar, inter-specimen temporal-bone CBCT | 3 | Dice | 0.150 | −0.332 | 0.289 | 0 | 116 |
- Reference
- Demir et al., WBIR 2024