VBoussot commited on
Commit
e00ba49
·
verified ·
1 Parent(s): b2ddacf

Update README

Browse files
Files changed (1) hide show
  1. README.md +9 -5
README.md CHANGED
@@ -11,16 +11,23 @@ tags:
11
 
12
  # 🧠 TorchScript Models for the IMPACT Semantic Similarity Metric
13
 
14
- This repository hosts multiple **TorchScript-exported pretrained models** used by the **IMPACT** similarity metric for semantic medical image registration.
15
 
16
- The IMPACT metric is described in the following preprint, currently under review:
17
 
18
  > **IMPACT: A Generic Semantic Loss for Multimodal Medical Image Registration**
19
  > *V. Boussot, C. Hémon, J.-C. Nunes, J. Downling, S. Rouzé, C. Lafond, A. Barateau, J.-L. Dillenseger*
20
  > [arXiv:2503.24121 [cs.CV]](https://arxiv.org/abs/2503.24121)
21
 
 
 
 
 
 
22
  ---
23
 
 
 
24
  The TorchScript models provided in this repository were exported from publicly available pretrained networks. These include:
25
 
26
  - **TotalSegmentator (TS)** — U-Net models trained for full-body anatomical segmentation
@@ -34,9 +41,6 @@ In addition, the repository also includes:
34
 
35
  - **MIND** — A handcrafted Modality Independent Neighborhood Descriptor, wrapped in TorchScript
36
 
37
- ---
38
-
39
- ## 📚 Pretrained Model References
40
 
41
  | Model | Specialization | Paper / Reference | Field of View | License |
42
  |----------------|----------------------------------------|-------------------------------------------------------------|------------------------|--------------|
 
11
 
12
  # 🧠 TorchScript Models for the IMPACT Semantic Similarity Metric
13
 
14
+ This repository provides a collection of **TorchScript-exported pretrained models** designed for use with the **IMPACT** similarity metric, enabling semantic medical image registration through feature-level comparison.
15
 
16
+ The IMPACT metric is introduced in the following preprint, currently under review:
17
 
18
  > **IMPACT: A Generic Semantic Loss for Multimodal Medical Image Registration**
19
  > *V. Boussot, C. Hémon, J.-C. Nunes, J. Downling, S. Rouzé, C. Lafond, A. Barateau, J.-L. Dillenseger*
20
  > [arXiv:2503.24121 [cs.CV]](https://arxiv.org/abs/2503.24121)
21
 
22
+ 🔧 The full implementation of IMPACT, along with its integration into the **Elastix** framework, is available in the repository:
23
+ ➡️ [github.com/vboussot/ImpactLoss](https://github.com/vboussot/ImpactLoss)
24
+
25
+ This repository also includes example parameter maps, TorchScript model handling utilities, and a ready-to-use Docker environment for quick experimentation and reproducibility.
26
+
27
  ---
28
 
29
+ ## 📚 Pretrained Model
30
+
31
  The TorchScript models provided in this repository were exported from publicly available pretrained networks. These include:
32
 
33
  - **TotalSegmentator (TS)** — U-Net models trained for full-body anatomical segmentation
 
41
 
42
  - **MIND** — A handcrafted Modality Independent Neighborhood Descriptor, wrapped in TorchScript
43
 
 
 
 
44
 
45
  | Model | Specialization | Paper / Reference | Field of View | License |
46
  |----------------|----------------------------------------|-------------------------------------------------------------|------------------------|--------------|