Added sample data for running the docker image, also added the script for generating the sample data. Zipped the docker image generated. Make changes to nifit_wrapper to use 4, 4

Author: Unique-Usman

.github/workflows/docker-build-and-run.yml

@@ -25,24 +25,27 @@ jobs:
         run: |
           docker build -t tf2.4_ivim-mri_codecollection -f Docker/Dockerfile .
 
-      - name: Save Docker image to a tarball
+      - name: Save and compress Docker image 
         run: |
-          docker save -o tf2.4_ivim-mri_codecollection.tar tf2.4_ivim-mri_codecollection
+          docker save tf2.4_ivim-mri_codecollection | gzip > tf2.4_ivim-mri_codecollection.tar.gz
 
       - name: Upload Docker image artifact
         uses: actions/upload-artifact@v4
         with:
           name: docker-image
-          path: tf2.4_ivim-mri_codecollection.tar
+          path: tf2.4_ivim-mri_codecollection.tar.gz
 
       - name: Run Docker container
         run: |
           docker run --rm --name TF2.4_IVIM-MRI_CodeCollection \
             -v ${{ github.workspace }}:/usr/src/app \
             -v ${{ github.workspace }}:/usr/app/output \
-            tf2.4_ivim-mri_codecollection brain.nii.gz brain.bvec brain.bval
+            tf2.4_ivim-mri_codecollection Docker/ivim_simulation.nii.gz Docker/ivim_simulation.bvec Docker/ivim_simulation.bval
 
-      - name: Clean up
+      - name: Clean up artifacts and Docker image
         run: |
-          docker rmi tf2.4_ivim-mri_codecollection
-          rm tf2.4_ivim-mri_codecollection.tar
+          docker rmi tf2.4_ivim-mri_codecollection || true
+          rm -f tf2.4_ivim-mri_codecollection.tar.gz
+          rm -f ${{ github.workspace }}/f.nii.gz
+          rm -f ${{ github.workspace }}/dp.nii.gz
+          rm -f ${{ github.workspace }}/d.nii.gz

Docker/generate_signal_docker_test.py

@@ -0,0 +1,31 @@
+import numpy as np
+import nibabel as nib
+from utilities.data_simulation.GenerateData import GenerateData
+
+def save_nii(data, filename='ivim_image.nii.gz'):
+    """
+    Save the data as a NIfTI file (.nii.gz)
+    """
+    nii_img = nib.Nifti1Image(data, affine=np.eye(4))
+    nib.save(nii_img, filename)
+
+# Set random seed for reproducibility
+np.random.seed(42)
+
+# Create GenerateData instance
+gd = GenerateData()
+
+# Generate random input data
+shape = (10, 10, 5)
+f_in = np.random.uniform(low=0, high=1, size=shape)
+D_in = np.random.uniform(low=0, high=1e-3, size=shape)
+Dp_in = np.random.uniform(low=0, high=1e-1, size=shape)
+S0 = 1000  # Setting a constant S0 for simplicity
+bvals = np.array([0, 50, 100, 500, 1000])
+bvals_reshaped = np.broadcast_to(bvals, shape)
+
+# Generate IVIM signal
+images = gd.ivim_signal(D_in, Dp_in, f_in, S0, bvals_reshaped)
+
+# Save the generated image as a NIfTI file
+save_nii(images)

Docker/ivim_simulation.bval

@@ -0,0 +1 @@
+0 50 100 500 1000

Docker/ivim_simulation.bvec

@@ -0,0 +1,3 @@
+1 0 0
+0 1 0
+0 0 1

Docker/ivim_simulation.nii.gz

@@ -56,7 +56,9 @@ def save_nifti_file(data, output_file, affine=None, **kwargs):
     For saving the 3d nifti images of the output of the algorithm
     """
     if affine is None:
-        affine = np.eye(data.ndim + 1)
+        affine = np.eye(4)
+    else:
+        affine = np.array(affine.reshape(4, 4))
     output_img = nib.nifti1.Nifti1Image(data, affine , **kwargs)
     nib.save(output_img, output_file)
 

WrapImage/nifti_wrapper.py

@@ -53,7 +53,7 @@ def exponential_signal(self, D, bvalues):
         bvalues : list or array of float
             The diffusion (b-values)
         """
-        assert D >= 0, 'D must be >= 0'
+        assert np.all(D >= 0), 'all values in D must be >= 0'
         return self._op.exp(-self._op.asarray(bvalues, dtype='float64') * D)
 
     def multiexponential_signal(self, D, F, S0, bvalues):
@@ -75,7 +75,7 @@ def multiexponential_signal(self, D, F, S0, bvalues):
         assert len(D) == len(F), 'D and F must be the same length'
         signal = self._op.zeros_like(bvalues)
         for [d, f] in zip(D, F):
-            signal += f * self.exponential_signal(d, bvalues)
+            signal += f * self.exponential_signal(d[0], bvalues)
         signal *= S0
         return signal
 
@@ -116,7 +116,7 @@ def linear_signal(self, D, bvalues, offset=0):
         offset : float
             The signal offset
         """
-        assert D >= 0, 'D must be >= 0'
+        assert np.all(D >= 0), 'every value in D must be >= 0'
         data = -D * np.asarray(bvalues)
         return data + offset
 
@@ -144,4 +144,4 @@ def multilinear_signal(self, D, F, S0, bvalues, offset=0):
             signal += f * self.linear_signal(d, bvalues)
         signal *= S0
         signal += offset
-        return signal
+        return signal