VBoussot commited on
Commit
564cff2
·
1 Parent(s): 2f7c1ac

Drop the placeholder checkpoint; these engines are weightless

Browse files

model.pt was a dummy 1249-byte checkpoint that only existed to satisfy the predictor's
'at least one checkpoint' rule. These registration engines have no trained weights, and
konfai now runs a weightless model (0 parameters) with no checkpoint, so each preset
declares 'models': [] and ships no .pt. Requires konfai >= 1.6.1.

CBCT_CT_HeadNeck/app.json CHANGED
@@ -1,13 +1,11 @@
1
  {
2
  "display_name": "CBCT/CT Head&Neck",
3
  "short_description": "Optimized preset for CBCT/CT registration on head & neck",
4
- "description": "A five-level recursive B-spline deformable registration optimized for CBCT/CT head-and-neck alignment, driven by the IMPACT metric using semantic features extracted from pretrained TotalSegmentator TorchScript models. The optimization follows a multi-resolution ASGD scheme with up to 300, 300, 200, 200, and 150 iterations and 2000 stochastic spatial samples per level. Features are extracted at progressively finer voxel scales (6 mm, 3 mm, 3 mm, 2\u00d72\u00d73 mm, 2\u00d72\u00d73 mm) using L1 distances on selected internal layers of the network. A composite objective (IMPACT + mutual information + bending energy penalty, with increased MI weight) ensures robust cross-modality alignment in complex head-and-neck anatomy while enforcing smooth, physically plausible deformations.",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
1
  {
2
  "display_name": "CBCT/CT Head&Neck",
3
  "short_description": "Optimized preset for CBCT/CT registration on head & neck",
4
+ "description": "A five-level recursive B-spline deformable registration optimized for CBCT/CT head-and-neck alignment, driven by the IMPACT metric using semantic features extracted from pretrained TotalSegmentator TorchScript models. The optimization follows a multi-resolution ASGD scheme with up to 300, 300, 200, 200, and 150 iterations and 2000 stochastic spatial samples per level. Features are extracted at progressively finer voxel scales (6 mm, 3 mm, 3 mm, 2×2×3 mm, 2×2×3 mm) using L1 distances on selected internal layers of the network. A composite objective (IMPACT + mutual information + bending energy penalty, with increased MI weight) ensures robust cross-modality alignment in complex head-and-neck anatomy while enforcing smooth, physically plausible deformations.",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
CBCT_CT_HeadNeck/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
CBCT_CT_MRSeg/app.json CHANGED
@@ -1,13 +1,11 @@
1
  {
2
  "display_name": "CBCT/CT preset with MRSegmentator",
3
  "short_description": "Generic CBCT/CT deformable registration using MRSegmentator features",
4
- "description": "A four-level recursive B-spline deformable registration optimized for generic CBCT/CT alignment, driven by the IMPACT metric using semantic features extracted from the pretrained MRSegmentator model. The scheme follows a multi-resolution strategy with up to 300, 300, 250, and 200 ASGD iterations and 2000 stochastic spatial samples per level. Features are extracted at progressively finer voxel scales (3 mm, 3 mm, 2\u00d72\u00d73 mm, 2\u00d72\u00d73 mm), with a level-dependent combination of Dice-based segmentation overlap and L1 feature distances on selected internal layers of MRSegmentator. Early levels rely on pure Dice supervision, while finer stages progressively integrate feature-level alignment with increasing L1 contribution (0.3/0.7, 0.5/0.5) and a final purely feature-based stage. The optimization minimizes a composite objective (IMPACT + mutual information + bending energy penalty), enabling robust cross-modality alignment between CBCT and CT while enforcing smooth, physically plausible deformations.",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
1
  {
2
  "display_name": "CBCT/CT preset with MRSegmentator",
3
  "short_description": "Generic CBCT/CT deformable registration using MRSegmentator features",
4
+ "description": "A four-level recursive B-spline deformable registration optimized for generic CBCT/CT alignment, driven by the IMPACT metric using semantic features extracted from the pretrained MRSegmentator model. The scheme follows a multi-resolution strategy with up to 300, 300, 250, and 200 ASGD iterations and 2000 stochastic spatial samples per level. Features are extracted at progressively finer voxel scales (3 mm, 3 mm, 2×2×3 mm, 2×2×3 mm), with a level-dependent combination of Dice-based segmentation overlap and L1 feature distances on selected internal layers of MRSegmentator. Early levels rely on pure Dice supervision, while finer stages progressively integrate feature-level alignment with increasing L1 contribution (0.3/0.7, 0.5/0.5) and a final purely feature-based stage. The optimization minimizes a composite objective (IMPACT + mutual information + bending energy penalty), enabling robust cross-modality alignment between CBCT and CT while enforcing smooth, physically plausible deformations.",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
CBCT_CT_MRSeg/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
ConvexAdam_Coarse/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
ConvexAdam_Coarse/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
ConvexAdam_Composite/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
ConvexAdam_Composite/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
ConvexAdam_Fine/app.json CHANGED
@@ -39,9 +39,7 @@
39
  "batch_size": 1
40
  }
41
  },
42
- "models": [
43
- "model.pt"
44
- ],
45
  "inputs": {
46
  "Fixed": {
47
  "display_name": "Fixed image",
 
39
  "batch_size": 1
40
  }
41
  },
42
+ "models": [],
 
 
43
  "inputs": {
44
  "Fixed": {
45
  "display_name": "Fixed image",
ConvexAdam_Fine/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
FireANTs_Affine/app.json CHANGED
@@ -1,13 +1,11 @@
1
  {
2
  "display_name": "FireANTs (Affine)",
3
  "short_description": "Rigid + Affine linear alignment on GPU (FireANTs), no deformable stage.",
4
- "description": "GPU linear registration with FireANTs: a Rigid (MI, centre-of-mass init) then Affine (MI) alignment optimised with Riemannian Adam, with no deformable stage. Produces the moved image and the affine displacement field on the fixed grid \u2014 use it on its own, or as a linear pre-align feeding a deformable preset. FireANTs is an external dependency under the FireANTs License v1.0 (see NOTICE).",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
1
  {
2
  "display_name": "FireANTs (Affine)",
3
  "short_description": "Rigid + Affine linear alignment on GPU (FireANTs), no deformable stage.",
4
+ "description": "GPU linear registration with FireANTs: a Rigid (MI, centre-of-mass init) then Affine (MI) alignment optimised with Riemannian Adam, with no deformable stage. Produces the moved image and the affine displacement field on the fixed grid — use it on its own, or as a linear pre-align feeding a deformable preset. FireANTs is an external dependency under the FireANTs License v1.0 (see NOTICE).",
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
FireANTs_Affine/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
FireANTs_IMPACT/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
FireANTs_IMPACT/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
FireANTs_SyN/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
FireANTs_SyN/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
Generic_Rigid/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
Generic_Rigid/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
Generic_Rigid_BSpline/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
Generic_Rigid_BSpline/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
MR_CT_HeadNeck/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
MR_CT_HeadNeck/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
MR_CT_MRSeg/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
MR_CT_MRSeg/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249
 
 
 
 
MR_CT_TS/app.json CHANGED
@@ -5,9 +5,7 @@
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
- "models": [
9
- "model.pt"
10
- ],
11
  "inputs": {
12
  "Fixed": {
13
  "display_name": "Fixed image",
 
5
  "task": "registration",
6
  "tta": 0,
7
  "mc_dropout": 0,
8
+ "models": [],
 
 
9
  "inputs": {
10
  "Fixed": {
11
  "display_name": "Fixed image",
MR_CT_TS/model.pt DELETED
@@ -1,3 +0,0 @@
1
- version https://git-lfs.github.com/spec/v1
2
- oid sha256:de99fbc36331ce674639acc774f52b4a2d0027f2f312d9d28669e831a0c4fd7e
3
- size 1249