DOC README update

Author: SebastianoF

README.md

@@ -32,14 +32,21 @@ segmentations in nifti format.
 
 ### Introductory example
 
-Given a segmentation `my_input_data/my_segm.nii.gz` you want to change the labels values from [1, 2, 3, 4, 5, 6] to [2, 12, 4, 7, 5, 6]. Then:
+Given a segmentation `my_input_folder_path/my_segm.nii.gz` imagine you want to change the labels values from [1, 2, 3, 4, 5, 6] to [2, 12, 4, 7, 5, 6]
+and save the result in `my_output_folder_path/my_new_segm.nii.gz`. Then:
 
 ```python
 import nilabels as nis
 
+# option 1
+nis_app = nis.App('my_input_folder_path', 'my_output_folder_path')
+nis_app.manipulate_labels.relabel('my_segm.nii.gz', 'my_new_segm.nii.gz', 
+                                  [1, 2, 3, 4, 5, 6], [2, 12, 4, 7, 5, 6])
 
-nis_app = nis.App('my_input_data')
-nis_app.manipulate.relabel('my_segm.nii.gz', 'my_new_segm.nii.gz', [1, 2, 3, 4, 5, 6], [2, 12, 4, 7, 5, 6])
+# option 2
+nis_app = nis.App()
+nis_app.manipulate_labels.relabel('my_input_folder_path/my_segm.nii.gz', 'my_output_folder_path/my_new_segm.nii.gz', 
+                                  [1, 2, 3, 4, 5, 6], [2, 12, 4, 7, 5, 6])
 ```
 
 ### How to install (in development mode) 

nilabels/agents/fuser.py

@@ -19,6 +19,8 @@ def seg_LabFusion(self, pfi_target, pfi_result, list_pfi_segmentations, list_pfi
                       prepare_data_only=False, seg_output_name='res_4d_seg', warp_output_name='res_4d_warp',
                       output_tag=''):
         """
+        Stack and fuse anatomical images and segmentations in a single command.
+        Based on NiftySeg if prepare_data_only is False.
         :param pfi_target: path to file to the target of the segmentation
         :param pfi_result: path to file where to store the result.
         :param list_pfi_segmentations: list of the segmentations to fuse

nilabels/agents/intensities_manipulator.py

@@ -21,7 +21,6 @@ def __init__(self, input_data_folder=None, output_data_folder=None, path_label_d
         self.pfo_out = output_data_folder
         self.path_label_descriptor = path_label_descriptor
 
-    # TODO embed path label descriptors.
     def normalise_below_label(self, filename_image_in, filename_image_out, filename_segm, labels, stats=np.median):
         """
 

nilabels/agents/measurer.py

@@ -139,7 +139,7 @@ def global_dist(self, segm_1_filename, segm_2_filename, where_to_save=None,
         return se_global_distances
 
     def topology(self):
-        # WIP: island detections, graph detections, cc detections from detector tools
+        # TODO: island detections, graph detections, cc detections from detector tools
         print('topology for {} is in the TODO list!'.format(self.__class__))
 
     def groupwise_global_measures_comparisons(self, list_path_A, list_path_B, pfo_where_to_save,

nilabels/tools/caliber/distances.py

@@ -2,7 +2,7 @@
 import pandas as pa
 from scipy import ndimage as nd
 
-from nilabels.tools.detections.contours import contour_from_array_at_label_l
+from nilabels.tools.detections.contours import contour_from_array_at_label
 
 
 # --- Auxiliaries
@@ -186,8 +186,8 @@ def symmetric_contour_distance_one_label(im1, im2, lab, return_mm3, formula='nor
     if np.count_nonzero(arr1) == 0 or np.count_nonzero(arr2) == 0:
         return np.nan
 
-    arr1_contour = contour_from_array_at_label_l(arr1, 1)
-    arr2_contour = contour_from_array_at_label_l(arr2, 1)
+    arr1_contour = contour_from_array_at_label(arr1, 1)
+    arr2_contour = contour_from_array_at_label(arr2, 1)
 
     if return_mm3:
         dtb1 = nd.distance_transform_edt(1 - arr1_contour, sampling=list(np.diag(im1.affine[:3, :3])))

nilabels/tools/defs.py

@@ -4,7 +4,7 @@
 root_dir = os.path.dirname(os.path.dirname(os.path.abspath(os.path.dirname(__file__))))
 
 info = {
-        "name": "LABelsToolkit",
+        "name": "NiLabels",
         "version": "0.1",
         "description": "",
         "repository": {

nilabels/tools/detections/contours.py

@@ -3,20 +3,20 @@
 from nilabels.tools.aux_methods.utils_nib import set_new_data
 
 
-def contour_from_array_at_label_l(im_arr, l, thr=0.3, omit_axis=None, verbose=0):
+def contour_from_array_at_label(im_arr, lab, thr=0.3, omit_axis=None, verbose=0):
     """
     Get the contour of a single label
     :param im_arr: input array with segmentation
-    :param l: considered label
+    :param lab: considered label
     :param thr: threshold (default 0.3) increase to increase the contour thickness.
     :param omit_axis: a directional axis preference for the contour creation, to avoid "walls" when scrolling
     the 3d image in a particular direction. None if no preference axis is expected.
     :param verbose:
     :return:
     """
     if verbose > 0:
-        print('eroding label {}'.format(l))
-    array_label_l = im_arr == l
+        print('eroding label {}'.format(lab))
+    array_label_l = im_arr == lab
     assert isinstance(array_label_l, np.ndarray)
     gra = np.gradient(array_label_l.astype(np.bool).astype(np.float64))
     if omit_axis is None:
@@ -46,6 +46,6 @@ def contour_from_segmentation(im_segm, omit_axis=None, verbose=0):
     output_arr = np.zeros_like(im_segm.get_data(), dtype=im_segm.get_data_dtype())
 
     for la in list_labels:
-        output_arr += contour_from_array_at_label_l(im_segm.get_data(), la, omit_axis=omit_axis, verbose=verbose)
+        output_arr += contour_from_array_at_label(im_segm.get_data(), la, omit_axis=omit_axis, verbose=verbose)
 
     return set_new_data(im_segm, output_arr.astype(np.bool) * im_segm.get_data(), new_dtype=im_segm.get_data_dtype())

nilabels/tools/image_colors_manipulations/cutter.py

@@ -49,6 +49,7 @@ def apply_a_mask_nib(im_input, im_mask):
     """
     assert len(im_mask.shape) == 3
 
+    # TODO correct this: make the cut_4d_volume_with_a_1_slice_mask merged hrer
     if not im_mask.shape == im_input.shape[:3]:
         msg = 'Provided mask and image does not have compatible dimension: {0} and {1}'.format(
             im_input.shape, im_mask.shape)

nilabels/tools/image_colors_manipulations/normaliser.py

@@ -74,7 +74,6 @@ def intensities_normalisation_linear(im_input, im_segm, im_mask_foreground=None,
     return set_new_data(im_input, im_mask_foreground_data * (a * im_input.get_data() + b))
 
 
-
 def mahalanobis_distance_map(im, im_mask=None, trim=False):
     """
     From an image to its Mahalanobis distance map