mrid-python 0.1.4__py3-none-any.whl → 0.1.5__py3-none-any.whl

mrid/preprocessing/hd_bet.py

@@ -235,7 +235,7 @@ def skullstrip_D(
 
     # optionally add with skullstripped postfix
     if keep_original:
-        skullstripped = {f"{k}_hd_bet": v for k,v in skullstripped}
+        skullstripped = {(f"{k}_hd_bet" if k != "seg_hd_bet" else k): v for k,v in skullstripped.items()}
         skullstripped.update(images.copy())
 
     return skullstripped

mrid/preprocessing/mask.py

@@ -1,6 +1,5 @@
 import SimpleITK as sitk
 import numpy as np
-import SimpleITK as sitk
 
 from ..loading.convert import ImageLike, tositk, tonumpy
 
@@ -15,12 +14,12 @@ def expand_binary_mask(binary_mask: ImageLike, expand: int) -> sitk.Image:
             Negative values dilate the mask by this many pixels.
     """
     binary_mask = tositk(binary_mask)
-    if expand > 0:
+    if expand < 0:
         inverted_mask = 1 - binary_mask
-        return 1 - sitk.BinaryDilate(inverted_mask, (expand, expand, expand))
+        return 1 - sitk.BinaryDilate(inverted_mask, (-expand, -expand, -expand))
 
-    if expand < 0:
-        return sitk.BinaryDilate(binary_mask, (-expand, -expand, -expand))
+    if expand > 0:
+        return sitk.BinaryDilate(binary_mask, (expand, expand, expand))
 
     return binary_mask
 

mrid/preprocessing/spatial.py

@@ -73,8 +73,9 @@ def resize(img: ImageLike, new_size: Sequence[int], interpolator=sitk.sitkLinear
 
     return sitk.Resample(img, reference_image, centered_transform, interpolator, 0.0)
 
-def downsample(image:ImageLike, factor:float, dims: Sequence[int] | None, interpolator=sitk.sitkLinear) -> sitk.Image:
+def downsample(image:ImageLike, factor:float, dims: int | Sequence[int] | None, interpolator=sitk.sitkLinear) -> sitk.Image:
     """factor = 2 for 2x downsampling"""
+    if isinstance(dims, int): dims = (dims, )
     image = tositk(image)
     size = sitk.GetArrayFromImage(image).shape
     size = [round(s/factor) if (dims is None or i in dims) else s for i,s in enumerate(size)]

mrid/preprocessing/synthstrip.py

@@ -70,6 +70,7 @@ def run_synthstrip(
     fill: int | None = None,
     no_csf: bool | None = None,
     model: str | os.PathLike | None = None,
+    verbose: bool = True,
 ):
     """Runs ``synthstrip`` command-line routine via ``subprocess.run``.
 
@@ -108,7 +109,10 @@ def run_synthstrip(
     if model is not None: command.extend(["--model", os.path.normpath(model)])
 
     # run
-    subprocess.run(command, check=True)
+    if verbose:
+        subprocess.run(command, check=True)
+    else:
+        subprocess.run(command, check=True, stdout=subprocess.DEVNULL, stderr=subprocess.STDOUT)
 
 def predict_brain_mask(
     synthstrip_script_path: str | os.PathLike,
@@ -117,6 +121,7 @@ def predict_brain_mask(
     border: int | None = None,
     threads: int | None = None,
     model: str | os.PathLike | None = None,
+    verbose: bool = True,
 ):
     """Returns brain mask of ``input`` predicted by ``synthstrip``.
 
@@ -141,6 +146,7 @@ def predict_brain_mask(
             border=border,
             threads=threads,
             model=model,
+            verbose=verbose,
         )
 
         brain_mask = tositk(os.path.join(tmpdir, "synthstrip_mask.nii.gz"))
@@ -155,6 +161,7 @@ def skullstrip(
     threads: int | None = None,
     model: str | os.PathLike | None = None,
     expand: int = 0,
+    verbose: bool = True,
 ):
     """Skullstrips ``input`` using synthstrip.
 
@@ -181,6 +188,7 @@ def skullstrip(
         border=border,
         threads=threads,
         model=model,
+        verbose=verbose
     )
     if expand != 0:
         mask = expand_binary_mask(mask, expand=expand)
@@ -202,6 +210,8 @@ def skullstrip_D(
     include_mask: bool = False,
     keep_original: bool = False,
 
+    verbose: bool = True,
+
 ) -> dict[str, sitk.Image]:
     """Predicts brain mask of ``images[key]`` using synthstrip, then uses this mask to skull strip all values in ``images``.
 
@@ -233,6 +243,7 @@ def skullstrip_D(
         border=border,
         threads=threads,
         model=model,
+        verbose=verbose,
     )
     skullstripped = {}
 
@@ -252,7 +263,7 @@ def skullstrip_D(
 
     # optionally add with skullstripped postfix
     if keep_original:
-        skullstripped = {f"{k}_synthstrip": v for k,v in skullstripped}
+        skullstripped = {(f"{k}_synthstrip" if k != "seg_synthstrip" else k): v for k,v in skullstripped.items()}
         skullstripped.update(images.copy())
 
     return skullstripped

mrid/study.py

@@ -79,6 +79,19 @@ class Study(UserDict[str, sitk.Image | Any]):
         study[key] = item
         return study
 
+    def remove(self, *keys: str | Sequence[str]):
+        """Returns a new study without specified keys"""
+        keys_proc = []
+        for k in keys:
+            if isinstance(k, str): keys_proc.append(k)
+            else: keys_proc.extend(k)
+
+        study = self.copy()
+        for k in keys_proc:
+            del study[k]
+
+        return study
+
     def get_scans(self):
         """Returns a new ``Study`` with segmentations and info removed."""
         return self.__class__({k:v for k,v in self.items() if not k.startswith(("seg", "info"))})
@@ -158,7 +171,7 @@ class Study(UserDict[str, sitk.Image | Any]):
         and uses that bounding box to crop all other images, including segmentations.
 
         Args:
-            key: The key of the image to use for finding the foreground bounding box.
+            key: The key of the image (scan or segmentation) to use for finding the foreground bounding box.
         """
         d = preprocessing.cropping.crop_bg_D(self.get_images(), key)
         return Study(**d, **self.get_info())
@@ -231,6 +244,8 @@ class Study(UserDict[str, sitk.Image | Any]):
 
         include_mask: bool = False,
         keep_original: bool = False,
+
+        verbose: bool = True,
     ):
         """Returns a new study with all scans skullstripped.
 
@@ -260,6 +275,7 @@ class Study(UserDict[str, sitk.Image | Any]):
             key=key,
             gpu=gpu, border=border, threads=threads, model=model,
             expand=expand, include_mask=include_mask, keep_original=keep_original,
+            verbose=verbose,
         )
         return Study(**d, **self.get_segmentations(), **self.get_info())
 
@@ -460,7 +476,6 @@ class Study(UserDict[str, sitk.Image | Any]):
         arr = tfm(self.numpy(key))
         return self.add(f'{key}{postfix}', arr, reference_key=key)
 
-
     def numpy(self, key: str):
         """returns ``study[key]`` converted to a numpy array."""
         return tonumpy(self[key])
@@ -527,7 +542,7 @@ class Study(UserDict[str, sitk.Image | Any]):
 
     def plot(self):
         from .utils.plotting import plot_study
-        plot_study(self.get_images().numpy_dict())
+        return plot_study(self.get_images().numpy_dict())
 
     def save(
         self,

mrid/training/slicer.py

@@ -1,10 +1,11 @@
+import os
 import random
 from functools import partial
 from typing import Any, Literal, cast
 from collections.abc import Callable, Sequence
 import torch
 
-from ..loading import ImageLike, totensor
+from ..loading import ImageLike, totensor, tonumpy
 
 
 class SliceSampler:
@@ -216,6 +217,7 @@ class SliceDataset(torch.utils.data.Dataset):
         randflip: bool = True,
         flatten: bool = True,
         repeat: int = 1,
+        tfm: Callable[[torch.Tensor, torch.Tensor], Any] | None = None,
     ):
         super().__init__()
         self._callables = [s.random_weighted_callable(
@@ -223,6 +225,8 @@ class SliceDataset(torch.utils.data.Dataset):
 
         self._repeat = repeat
 
+        self.tfm = tfm
+
     def __len__(self):
         return len(self._callables) * self._repeat
 
@@ -232,4 +236,9 @@ class SliceDataset(torch.utils.data.Dataset):
         if i >= length:
             raise IndexError(f"Index {i} is larger than length of SliceDataset {length}")
 
-        return self._callables[i % len(self._callables)]()
+        img, seg = self._callables[i % len(self._callables)]()
+
+        if self.tfm is not None:
+            return self.tfm(img, seg)
+
+        return img, seg

mrid/training/transforms.py

@@ -9,29 +9,41 @@ ARR = TypeVar("ARR", bound=Any)
 def crop(
     arr: ARR,
     reduction: Sequence[int],
-    where: Literal["start", "end", "center"] = "center",
+    where: Literal["start", "end", "center", "random"] = "center",
 ) -> ARR:
-    """Crop ``input`` such that ``output.shape[i] = input.shape[i] - reduction[i]``"""
+    """Crop ``arr`` such that ``output.shape[i] = input.shape[i] - reduction[i]``"""
 
-    # create slices
-    if where == 'center':
-        slices = [(int(i / 2), -int(i / 2)) if i % 2 == 0 else (int(i / 2), -int(i / 2) - 1) for i in reduction]
+    shape = arr.shape[-len(reduction):]
+    slices = []
 
-    elif where == 'start':
-        slices = [(None, -i) for i in reduction]
+    for r, sh in zip(reduction, shape):
+        if r == 0:
+            slices.append(slice(None))
+            continue
 
-    elif where == 'end':
-        slices = [(i, None) for i in reduction]
+        if r < 0: raise ValueError("Reduction cannot be negative")
+        if r > sh: raise ValueError(f"Reduction {r} exceeds dimension size {sh}")
 
-    slices = [slice(i if i!=0 else None, j if j != 0 else None) for i, j in slices]
+        if where == 'start': start, end = 0, sh - r
+        elif where == 'end': start, end = r, sh
+        elif where == 'center':
+            start = r // 2
+            end = start + (sh - r)
+        elif where == 'random':
+            start = random.randint(0, r)
+            end = start + (sh - r)
+        else:
+            raise ValueError(f"Invalid where: {where}")
 
-    # crop with broadcasting
+        slices.append(slice(start, end))
+
+    # apply with broadcasting
     return arr[(..., *slices)]
 
 def crop_to_shape(
     input: ARR,
     shape: Sequence[int],
-    where: Literal["start", "end", "center"] = "center",
+    where: Literal["start", "end", "center", "random"] = "center",
 ) -> ARR:
     """Crop ``input`` to ``shape``."""
 
@@ -42,7 +54,6 @@ def crop_to_shape(
     return crop(input, [i - j for i, j in zip(input.shape, shape)], where=where)
 
 
-
 def shuffle_channels(x:torch.Tensor):
     """Shuffle first axis in a ``(C, *)`` tensor"""
     return x[torch.randperm(x.shape[0])]

mrid/utils/plotting.py

@@ -1,11 +1,12 @@
 from collections.abc import Mapping
 
-import matplotlib.gridspec as gridspec
-import matplotlib.pyplot as plt
 import numpy as np
 from ..loading import ImageLike, tonumpy
 
 def plot_study(data: Mapping[str, ImageLike]):
+    import matplotlib.gridspec as gridspec
+    import matplotlib.pyplot as plt
+
     data = {k: tonumpy(v) for k,v in data.items()}
     n_vals = len(data)
 
@@ -77,7 +78,8 @@ def plot_study(data: Mapping[str, ImageLike]):
         fig.text(box.x0 + box.width/2, box.y1 + 0.04, modality_name,
                  ha='center', va='bottom', fontsize=14, fontweight='bold')
 
-    plt.show()
+    # plt.show()
+    return fig
 
 if __name__ == "__main__":
 
@@ -97,8 +99,8 @@ if __name__ == "__main__":
         return mask
 
     plot_study({
-        "T1 Weighted (Sphere)": dummy_data((60, 60, 60), 'sphere'),
-        "T2 Weighted (Cube)": dummy_data((60, 60, 60), 'cube'),
-        "Proton Density (Noise)": dummy_data((60, 60, 60), 'noise'),
-        "FLAIR (Sphere)": dummy_data((60, 60, 60), 'sphere'),
+        "Sphere": dummy_data((60, 60, 60), 'sphere'),
+        "Cube": dummy_data((60, 60, 60), 'cube'),
+        "Noise": dummy_data((60, 60, 60), 'noise'),
+        "Another sphere": dummy_data((60, 60, 60), 'sphere'),
     })

mrid/utils/stl_utils.py

@@ -6,6 +6,69 @@ import SimpleITK as sitk
 
 from ..loading.convert import tositk
 
+# def stl2sitk(
+#     stl_path: str | os.PathLike,
+#     reference: str | os.PathLike | sitk.Image,
+#     fix_holes: bool = False,
+# ):
+#     """
+#     Loads an STL file under ``stl_path`` and converts it to ``sitk.Image`` aligned with ``reference``.
+#     Note that this might take a few minutes.
+
+#     Args:
+#         stl_path (str): Path to the STL segmentation file. The STL coordinates
+#                         MUST be in the same coordinate system as the ``reference``.
+#         reference (str | os.PathLike | sitk.Image): path to a directory of DICOM files or a NIfTI file, or a ``sitk.Image``.
+#         fix_holes (bool, optional): whether to try to fix holes in STL if they are detected (this can be very slow).
+#     """
+#     import trimesh
+
+#     # ------------------------------ load reference ------------------------------ #
+#     reference = tositk(reference)
+
+#     origin = np.array(reference.GetOrigin())
+#     spacing = np.array(reference.GetSpacing())
+#     # ct_direction = np.array(ct_image.GetDirection()).reshape(3, 3)
+#     size = np.array(reference.GetSize()) # Order: x, y, z
+#     shape_xyz = size
+#     shape_zyx = size[::-1]
+
+#     # --------------------------------- load STL --------------------------------- #
+#     mesh = trimesh.load_mesh(stl_path)
+
+#     if not mesh.is_watertight:
+#         warnings.warn(f"Warning: STL mesh '{stl_path}' is not watertight. Voxelization using 'contains' might be inaccurate.")
+#         if fix_holes:
+#             mesh.fill_holes()
+#             if not mesh.is_watertight:
+#                 warnings.warn("Warning: Failed to make mesh watertight after filling holes.")
+
+#     # ------------------------------- voxelize STL ------------------------------- #
+#     x_coords = origin[0] + np.arange(shape_xyz[0]) * spacing[0]
+#     y_coords = origin[1] + np.arange(shape_xyz[1]) * spacing[1]
+#     z_coords = origin[2] + np.arange(shape_xyz[2]) * spacing[2]
+
+#     # Use meshgrid to create a grid of coordinates
+#     # Note the 'ij' indexing to match the z, y, x array structure
+#     zz, yy, xx = np.meshgrid(z_coords, y_coords, x_coords, indexing='ij')
+
+#     # Stack coordinates into a (N, 3) array where N = Z*Y*X
+#     voxel_centers_xyz = np.stack([xx.ravel(), yy.ravel(), zz.ravel()], axis=-1)
+
+#     # This checks which voxel center points fall inside the mesh volume
+#     voxel_mask_flat = mesh.contains(voxel_centers_xyz)
+
+#     # Reshape the flat boolean mask back into the 3D CT shape (z, y, x)
+#     stl_array = voxel_mask_flat.reshape(shape_zyx).astype(np.uint8) # Use uint8 for masks
+
+#     # ------------------------------ make sitk.Image ----------------------------- #
+#     stl_sitk = sitk.GetImageFromArray(stl_array)
+
+#     stl_sitk.SetOrigin(reference.GetOrigin())
+#     stl_sitk.SetSpacing(reference.GetSpacing())
+#     stl_sitk.SetDirection(reference.GetDirection())
+#     return stl_sitk
+
 def stl2sitk(
     stl_path: str | os.PathLike,
     reference: str | os.PathLike | sitk.Image,
@@ -25,11 +88,10 @@ def stl2sitk(
 
     # ------------------------------ load reference ------------------------------ #
     reference = tositk(reference)
-
     origin = np.array(reference.GetOrigin())
     spacing = np.array(reference.GetSpacing())
-    # ct_direction = np.array(ct_image.GetDirection()).reshape(3, 3)
-    size = np.array(reference.GetSize()) # Order: x, y, z
+    size = np.array(reference.GetSize())  # x, y, z
+
     shape_xyz = size
     shape_zyx = size[::-1]
 
@@ -37,34 +99,44 @@ def stl2sitk(
     mesh = trimesh.load_mesh(stl_path)
 
     if not mesh.is_watertight:
-        warnings.warn(f"Warning: STL mesh '{stl_path}' is not watertight. Voxelization using 'contains' might be inaccurate.")
+        warnings.warn(f"STL mesh '{stl_path}' is not watertight.")
         if fix_holes:
             mesh.fill_holes()
-            if not mesh.is_watertight:
-                print("Warning: Failed to make mesh watertight after filling holes.")
 
-    # ------------------------------- voxelize STL ------------------------------- #
+    # ----------------------- find bounding box of the mask ---------------------- #
+    bounds = mesh.bounds  # [[min_x, min_y, min_z], [max_x, max_y, max_z]]
+    voxel_min = np.floor((bounds[0] - origin) / spacing).astype(int)
+    voxel_max = np.ceil((bounds[1] - origin) / spacing).astype(int)
+
+    z_min, z_max = np.clip([voxel_min[2], voxel_max[2]], 0, shape_xyz[2] - 1)
+
+    # --------------------------------- voxelize --------------------------------- #
+    stl_array = np.zeros(shape_zyx, dtype=np.uint8)
     x_coords = origin[0] + np.arange(shape_xyz[0]) * spacing[0]
     y_coords = origin[1] + np.arange(shape_xyz[1]) * spacing[1]
-    z_coords = origin[2] + np.arange(shape_xyz[2]) * spacing[2]
 
-    # Use meshgrid to create a grid of coordinates
-    # Note the 'ij' indexing to match the z, y, x array structure
-    zz, yy, xx = np.meshgrid(z_coords, y_coords, x_coords, indexing='ij')
+    yy, xx = np.meshgrid(y_coords, x_coords, indexing='ij')
+    points_2d = np.stack([xx.ravel(), yy.ravel()], axis=-1)
+
+    for z_idx in range(z_min, z_max + 1):
+        z_val = origin[2] + z_idx * spacing[2]
 
-    # Stack coordinates into a (N, 3) array where N = Z*Y*X
-    voxel_centers_xyz = np.stack([xx.ravel(), yy.ravel(), zz.ravel()], axis=-1)
+        # Create 3D points for this slice: [X, Y, current_Z]
+        points_3d = np.column_stack([
+            points_2d,
+            np.full(points_2d.shape[0], z_val)
+        ])
 
-    # This checks which voxel center points fall inside the mesh volume
-    voxel_mask_flat = mesh.contains(voxel_centers_xyz)
+        # Check containment for this slice only
+        mask_flat = mesh.contains(points_3d)
 
-    # Reshape the flat boolean mask back into the 3D CT shape (z, y, x)
-    stl_array = voxel_mask_flat.reshape(shape_zyx).astype(np.uint8) # Use uint8 for masks
+        # Reshape and insert into the 3D array
+        stl_array[z_idx, :, :] = mask_flat.reshape(shape_xyz[1], shape_xyz[0])
 
     # ------------------------------ make sitk.Image ----------------------------- #
     stl_sitk = sitk.GetImageFromArray(stl_array)
-
     stl_sitk.SetOrigin(reference.GetOrigin())
     stl_sitk.SetSpacing(reference.GetSpacing())
     stl_sitk.SetDirection(reference.GetDirection())
+
     return stl_sitk

{mrid_python-0.1.4.dist-info → mrid_python-0.1.5.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mrid-python
-Version: 0.1.4
+Version: 0.1.5
 Summary: Tools for working with 3D medical images and segmentations - registration, brain skull-stripping, etc.
 Author-email: Ivan Nikishev <nkshv2@gmail.com>
 Project-URL: Homepage, https://github.com/inikishev/mrid
@@ -16,7 +16,7 @@ Requires-Dist: SimpleITK
 
 mrid is a library for preprocessing of 3D images, particularly medical images.
 
-It provide interfaces for many medical image processing tools such as [SimpleElastix](https://simpleelastix.github.io/), [HD-BET](https://github.com/MIC-DKFZ/HD-BET#Installation), [SynthStrip](https://surfer.nmr.mgh.harvard.edu/docs/synthstrip/), [CTSeg](https://github.com/WCHN/CTseg). Note that those libraries are not bundled with mrid, I've included installation instructions in all notebooks.
+It provide interfaces for many medical image processing tools such as [SimpleElastix](https://simpleelastix.github.io/), [HD-BET](https://github.com/MIC-DKFZ/HD-BET#Installation), [SynthStrip](https://surfer.nmr.mgh.harvard.edu/docs/synthstrip/), [CTSeg](https://github.com/WCHN/CTseg). Note that those libraries are not bundled, installation instructions are included in all examples below.
 
 ### Installation
 
@@ -32,10 +32,6 @@ or
 pip install git+https://github.com/inikishev/mrid
 ```
 
-### Basics
-
-The images you pass to all functions in mrid can be path to a .nii.gz file, DICOM directory, sitk.Image, numpy array or torch tensor. All functions return results as `sitk.Image`. If you need a numpy array, you can use `mrid.tonumpy(sitk_image)`.
-
 ### Registering images with SimpleITK-SimpleElastix
 
 [SimpleElastix](https://simpleelastix.github.io/) is a robust tool for image registration which works really well out-of-the-box. It works on both Windows and Linux.
@@ -69,7 +65,7 @@ Many [BraTS](https://www.synapse.org/brats) datasets are provided as skullstripp
 
 <img width="828" height="849" alt="image" src="https://github.com/user-attachments/assets/f1b38db3-6648-4660-a381-d68a2eb8508d" />
 
-(T1n image looks weird because that's just how it is in the dataset)
+(T1n image looks weird because that's just how it is in the zenodo dataset)
 
 ### References
 The MRIs for all images above are from https://zenodo.org/records/7213153.

{mrid_python-0.1.4.dist-info → mrid_python-0.1.5.dist-info}/RECORD

@@ -1,5 +1,5 @@
 mrid/__init__.py,sha256=d-h6l6e-sfPUKDo27_dXcx8IRdZ5ByDC1YtdkM6y39o,138
-mrid/study.py,sha256=oLvpsPCAbm0bhofl5wt4ekRedZo_59v5sTIEPhJZlKw,29084
+mrid/study.py,sha256=32dZe3x40lbKvIFsuQ8xaPJLEKElM-wGUv5Sm56J13g,29524
 mrid/atlas/__init__.py,sha256=ecOAVXUOt0TUI_wK19Ed05arF6b_G61wucVq2c8U4ZY,110
 mrid/atlas/MNI152/__init__.py,sha256=XTakiwb_rFxpO08O45tVKyHDzcT3xvJoLLMLu5EJhGw,3038
 mrid/atlas/SRI24/__init__.py,sha256=jFFY6LeJcWwCtxVJGbdWPEUGFKZjqzxu01ZTbbSAwEo,2939
@@ -9,26 +9,26 @@ mrid/preprocessing/CTseg.py,sha256=Zum1bvvU28JptrXLIOAaWVQA6LWPfaheCo-6liN1Gq4,3
 mrid/preprocessing/__init__.py,sha256=UNkuK_OKqBSdAt7n8dc7t5p7206W2wwksJwGy7rXdr0,412
 mrid/preprocessing/bias_field_correction.py,sha256=K2w75JkVIV_Gj7W-9gpHUwz2teEvgSRt5wX-CqQeRik,1143
 mrid/preprocessing/cropping.py,sha256=I26hxo7RuWqdjx8ApJ2t9FBPdi3PtRsBdMlLC2KgKTk,1336
-mrid/preprocessing/hd_bet.py,sha256=W7jsWVs7ZdoDBZe_OvSJa6zMOB9hpNK1_7SM_9s9VOY,11231
-mrid/preprocessing/mask.py,sha256=oD0jONMHlSpOBLuvzF35nc_cZ30L9q8cCIELBmAlffE,1375
+mrid/preprocessing/hd_bet.py,sha256=ybjtKfrdI5sFkjPZHh1rTPaMPy8JA7cIR71UlElKZ2A,11269
+mrid/preprocessing/mask.py,sha256=-x-9-CMbUR7Jx8p7YVtBb3r-OnGut9YefmEVI04jyDc,1350
 mrid/preprocessing/simple_elastix.py,sha256=gddw-B-IpGIyhfqj05kex8Zqb3_gMD8Z8w0BiLlx8dA,7975
-mrid/preprocessing/spatial.py,sha256=lV9XVf3jrHGJfRcEvgImlg6Rxb42WAZAYb0QLKgBpbA,4264
-mrid/preprocessing/synthstrip.py,sha256=fS3BESkOS9Du--zxBtKbqrt5e12WSyWEVnP7KBlJuk0,10149
+mrid/preprocessing/spatial.py,sha256=hsbGprUEcbeH0FY0_cgHZHjDOXGF55-ojtznuHttLkU,4316
+mrid/preprocessing/synthstrip.py,sha256=TaRFJn4vPmZ7V-SmKdaOTR3yZakoA3OfhwdjiBQg2Xs,10501
 mrid/training/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-mrid/training/slicer.py,sha256=mbis5a9D4wBCCxN1Gp9WAFnusxfMmiCq0Y_f04yi6y4,10239
-mrid/training/transforms.py,sha256=ZVpvk-19DfE6-PjgWqwaDcyI3_zPrN5DgiHI1644oMM,3641
+mrid/training/slicer.py,sha256=7eRQMDQ40FvpHs2J_eg_VoMnI42cOXG_tmgTDtYvVx8,10454
+mrid/training/transforms.py,sha256=fWoXH3KbkN3s-_A3h0bqt2yz9Ek5K8PzSIBaz3s5goM,4005
 mrid/utils/__init__.py,sha256=vDuaVk8uKs84R_Fg4RmGhQ-qHmpzS4--wPM86opvgig,153
 mrid/utils/dcm2niix.py,sha256=KHU7dgUVsU7WOEa82HFO-G0SsGIxYjeUFEBcX27mkWQ,4124
 mrid/utils/dicom_uid_fixer.py,sha256=xjjfvAF13Ni5CEgwPcr7oDT7excdEoeRGQetvjGkSRM,4046
-mrid/utils/plotting.py,sha256=UeEtMRZapJXKT27_567sELGgYZjJDsr-gPl-A6Xb33A,3904
+mrid/utils/plotting.py,sha256=bALTvNQxGXWl0fxGdKbSON47dz4OXfxVYEn6WeqdNK4,3885
 mrid/utils/python_utils.py,sha256=IYB8-KN4d6l-5ILgUNYux0vNrK0D0vh0AbSRqWhs-fo,1793
-mrid/utils/stl_utils.py,sha256=T4qT11NgvFCERfaB3duA_O1Qi9FBW7-b3B5aZKS5UU4,2888
+mrid/utils/stl_utils.py,sha256=vR6CqG1D4cpDmjkcZU0H5jCMVROeLwOTAzvoQzipgxs,5869
 mrid/utils/torch_utils.py,sha256=nKby_tPWzCGdEASkDUq7xNo-iUe9zjwmhq_chV0RCGU,407
 tests/test_loading.py,sha256=CciTvnqQ7hY-Vk2GgH2SWSu8nyS8lLeBqHu3XzClBLY,2141
 tests/test_preprocessing.py,sha256=HcoubITfmlJkgPaBbkxZiY04fJv9URBGMWDJWivDqLU,1300
 tests/test_study.py,sha256=pK__azqnMz8DSyOYaWQmBkpAqLntCIIjEz0m8oBS4RQ,3984
 tests/test_utils.py,sha256=Q4uf5dog673RDofOC6RPLYnxVN5NAIVRTiyTbf9Az74,359
-mrid_python-0.1.4.dist-info/METADATA,sha256=hxI4NAcTSuESvPcm__RfsqQKi5pi4nQS38tO0Ko94fI,4054
-mrid_python-0.1.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-mrid_python-0.1.4.dist-info/top_level.txt,sha256=lBv75ms7UoIM4elDVX3CbKiSlh-X0vABtaUwj4TGx4o,11
-mrid_python-0.1.4.dist-info/RECORD,,
+mrid_python-0.1.5.dist-info/METADATA,sha256=F5EwkIBp-KDPbLvwuMFWbsE-S-AUMk4OkuoK9u0jkm0,3795
+mrid_python-0.1.5.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+mrid_python-0.1.5.dist-info/top_level.txt,sha256=lBv75ms7UoIM4elDVX3CbKiSlh-X0vABtaUwj4TGx4o,11
+mrid_python-0.1.5.dist-info/RECORD,,