zaccharieramzi commited on
Commit
6b1036b
1 Parent(s): e57c8bb

added model card and model weights

Browse files
Files changed (2) hide show
  1. README.md +84 -0
  2. model_weights.h5 +3 -0
README.md ADDED
@@ -0,0 +1,84 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # NCPDNet-singlecoil-radial
2
+ ---
3
+ tags:
4
+ - TensorFlow
5
+ - MRI reconstruction
6
+ - MRI
7
+ datasets:
8
+ - fastMRI
9
+ ---
10
+
11
+ This is a non-Cartesian MRI reconstruction model for radial trajectories at acceleration factor 4.
12
+ The model uses 10 iterations and a small vanilla CNN.
13
+
14
+ ## Model description
15
+ For more details, see https://hal.inria.fr/hal-03188997.
16
+ This section is WIP.
17
+
18
+ ## Intended uses and limitations
19
+ This model can be used to reconstruct knee data from Siemens scanner at acceleration factor 4 in a radial acquisition setting.
20
+
21
+ ## How to use
22
+ This model can be loaded using the following repo: https://github.com/zaccharieramzi/fastmri-reproducible-benchmark.
23
+ After cloning the repo, `git clone https://github.com/zaccharieramzi/fastmri-reproducible-benchmark`, you can install the package via `pip install fastmri-reproducible-benchmark`.
24
+ The framework is TensorFlow.
25
+
26
+ You can initialize and load the model weights as follows:
27
+ ```python
28
+ import tensorflow as tf
29
+
30
+ from fastmri_recon.models.subclassed_models.ncpdnet import NCPDNet
31
+
32
+
33
+ model = NCPDNet(
34
+ im_size=(640, 400),
35
+ dcomp=True,
36
+ )
37
+ kspace_shape = 1
38
+ inputs = [
39
+ tf.zeros([1, 1, kspace_shape, 1], dtype=tf.complex64),
40
+ tf.zeros([1, 2, kspace_shape], dtype=tf.float32),
41
+ (tf.constant([320]), tf.ones([1, kspace_shape], dtype=tf.float32)),
42
+ ]
43
+ model(inputs)
44
+ model.load_weights('model_weights.h5')
45
+ ```
46
+
47
+ Using the model is then as simple as:
48
+ ```python
49
+ model([
50
+ kspace, # shape: [n_slices, 1, n_kspace_samples, 1]
51
+ traj, # shape: [n_slices, 1, 2, n_kspace_samples]
52
+ (
53
+ output_shape, # shape: [n_slices, 1]
54
+ dcomp, # shape: [n_slices, n_kspace_samples]
55
+ )
56
+ ])
57
+ ```
58
+
59
+ ## Limitations and bias
60
+ The limitations and bias of this model have not been properly investigated.
61
+
62
+ ## Training data
63
+ This model was trained using the [fastMRI dataset](https://fastmri.org/dataset/).
64
+
65
+ ## Training procedure
66
+ The training procedure is described in https://hal.inria.fr/hal-03188997.
67
+ This section is WIP.
68
+
69
+ ## Evaluation results
70
+ On the fastMRI validation dataset:
71
+ - PSNR: 32.66
72
+ - SSIM: 0.7327
73
+
74
+
75
+ ## Bibtex entry
76
+ ```
77
+ @unpublished{ramzi:hal-03188997,
78
+ TITLE = {{NC-PDNet: a Density-Compensated Unrolled Network for 2D and 3D non-Cartesian MRI Reconstruction}},
79
+ AUTHOR = {Ramzi, Zaccharie and G R, Chaithya and Starck, Jean-Luc and Ciuciu, Philippe},
80
+ YEAR = {2021},
81
+ MONTH = Sep,
82
+ }
83
+ ```
84
+
model_weights.h5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6f36293e0535c9722781cacfe71dcceda925b5ea23763a398921fb8e504e55d
3
+ size 716248