monai-weekly 1.4.dev2424__tar.gz → 1.4.dev2426__tar.gz

{monai_weekly-1.4.dev2424/monai_weekly.egg-info → monai_weekly-1.4.dev2426}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: monai-weekly
-Version: 1.4.dev2424
+Version: 1.4.dev2426
 Summary: AI Toolkit for Healthcare Imaging
 Home-page: https://monai.io/
 Author: MONAI Consortium
@@ -70,6 +70,7 @@ Requires-Dist: zarr; extra == "all"
 Requires-Dist: lpips==0.1.4; extra == "all"
 Requires-Dist: nvidia-ml-py; extra == "all"
 Requires-Dist: huggingface_hub; extra == "all"
+Requires-Dist: pyamg>=5.0.0; extra == "all"
 Provides-Extra: nibabel
 Requires-Dist: nibabel; extra == "nibabel"
 Provides-Extra: ninja
@@ -145,6 +146,8 @@ Provides-Extra: pynvml
 Requires-Dist: nvidia-ml-py; extra == "pynvml"
 Provides-Extra: huggingface-hub
 Requires-Dist: huggingface_hub; extra == "huggingface-hub"
+Provides-Extra: pyamg
+Requires-Dist: pyamg>=5.0.0; extra == "pyamg"
 
 <p align="center">
   <img src="https://raw.githubusercontent.com/Project-MONAI/MONAI/dev/docs/images/MONAI-logo-color.png" width="50%" alt='project-monai'>

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/__init__.py

@@ -93,4 +93,4 @@ except BaseException:
 
     if MONAIEnvVars.debug():
         raise
-__commit_id__ = "7449c4d23137e6fd264b35cc71c5396e2fce03c7"
+__commit_id__ = "2f62b81f411975f7622d0b137e18470c96b31178"

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/_version.py

@@ -8,11 +8,11 @@ import json
 
 version_json = '''
 {
- "date": "2024-06-16T02:17:58+0000",
+ "date": "2024-06-30T02:18:33+0000",
  "dirty": false,
  "error": null,
- "full-revisionid": "87d78f0c910874e98201d32532c2fd614fa902c6",
- "version": "1.4.dev2424"
+ "full-revisionid": "d622a16f927841fdd7d057b7553805405f0805e4",
+ "version": "1.4.dev2426"
 }
 '''  # END VERSION_JSON
 

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/apps/deepedit/transforms.py

@@ -30,7 +30,7 @@ measure, _ = optional_import("skimage.measure", "0.14.2", min_version)
 
 logger = logging.getLogger(__name__)
 
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 
 class DiscardAddGuidanced(MapTransform):

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/apps/deepgrow/transforms.py

@@ -27,7 +27,7 @@ from monai.utils import InterpolateMode, ensure_tuple, ensure_tuple_rep, min_ver
 from monai.utils.enums import PostFix
 
 measure, _ = optional_import("skimage.measure", "0.14.2", min_version)
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 DEFAULT_POST_FIX = PostFix.meta()
 

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/apps/nuclick/transforms.py

@@ -24,7 +24,7 @@ from monai.utils import StrEnum, convert_to_numpy, optional_import
 
 measure, _ = optional_import("skimage.measure")
 morphology, _ = optional_import("skimage.morphology")
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 
 class NuclickKeys(StrEnum):

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/apps/pathology/transforms/post/array.py

@@ -33,7 +33,7 @@ from monai.utils import TransformBackends, convert_to_numpy, optional_import
 from monai.utils.misc import ensure_tuple_rep
 from monai.utils.type_conversion import convert_to_dst_type, convert_to_tensor
 
-label, _ = optional_import("scipy.ndimage.measurements", name="label")
+label, _ = optional_import("scipy.ndimage", name="label")
 disk, _ = optional_import("skimage.morphology", name="disk")
 opening, _ = optional_import("skimage.morphology", name="opening")
 watershed, _ = optional_import("skimage.segmentation", name="watershed")

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/apps/pathology/utils.py

@@ -33,10 +33,10 @@ def compute_multi_instance_mask(mask: np.ndarray, threshold: float) -> Any:
     """
 
     neg = 255 - mask * 255
-    distance = ndimage.morphology.distance_transform_edt(neg)
+    distance = ndimage.distance_transform_edt(neg)
     binary = distance < threshold
 
-    filled_image = ndimage.morphology.binary_fill_holes(binary)
+    filled_image = ndimage.binary_fill_holes(binary)
     multi_instance_mask = measure.label(filled_image, connectivity=2)
 
     return multi_instance_mask

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/data/ultrasound_confidence_map.py

@@ -21,7 +21,9 @@ __all__ = ["UltrasoundConfidenceMap"]
 cv2, _ = optional_import("cv2")
 csc_matrix, _ = optional_import("scipy.sparse", "1.7.1", min_version, "csc_matrix")
 spsolve, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "spsolve")
+cg, _ = optional_import("scipy.sparse.linalg", "1.7.1", min_version, "cg")
 hilbert, _ = optional_import("scipy.signal", "1.7.1", min_version, "hilbert")
+ruge_stuben_solver, _ = optional_import("pyamg", "5.0.0", min_version, "ruge_stuben_solver")
 
 
 class UltrasoundConfidenceMap:
@@ -30,6 +32,9 @@ class UltrasoundConfidenceMap:
     It generates a confidence map by setting source and sink points in the image and computing the probability
     for random walks to reach the source for each pixel.
 
+    The official code is available at:
+    https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
     Args:
         alpha (float, optional): Alpha parameter. Defaults to 2.0.
         beta (float, optional): Beta parameter. Defaults to 90.0.
@@ -37,15 +42,33 @@ class UltrasoundConfidenceMap:
         mode (str, optional): 'RF' or 'B' mode data. Defaults to 'B'.
         sink_mode (str, optional): Sink mode. Defaults to 'all'. If 'mask' is selected, a mask must be when calling
             the transform. Can be 'all', 'mid', 'min', or 'mask'.
+        use_cg (bool, optional): Use Conjugate Gradient method for solving the linear system. Defaults to False.
+        cg_tol (float, optional): Tolerance for the Conjugate Gradient method. Defaults to 1e-6.
+            Will be used only if `use_cg` is True.
+        cg_maxiter (int, optional): Maximum number of iterations for the Conjugate Gradient method. Defaults to 200.
+            Will be used only if `use_cg` is True.
     """
 
-    def __init__(self, alpha: float = 2.0, beta: float = 90.0, gamma: float = 0.05, mode="B", sink_mode="all"):
+    def __init__(
+        self,
+        alpha: float = 2.0,
+        beta: float = 90.0,
+        gamma: float = 0.05,
+        mode="B",
+        sink_mode="all",
+        use_cg=False,
+        cg_tol=1e-6,
+        cg_maxiter=200,
+    ):
         # The hyperparameters for confidence map estimation
         self.alpha = alpha
         self.beta = beta
         self.gamma = gamma
         self.mode = mode
         self.sink_mode = sink_mode
+        self.use_cg = use_cg
+        self.cg_tol = cg_tol
+        self.cg_maxiter = cg_maxiter
 
         # The precision to use for all computations
         self.eps = np.finfo("float64").eps
@@ -228,17 +251,18 @@ class UltrasoundConfidenceMap:
         s = self.normalize(s)
 
         # Horizontal penalty
-        s[:vertical_end] += gamma
-        # s[vertical_end:diagonal_end] += gamma * np.sqrt(2) # --> In the paper it is sqrt(2)
-        # since the diagonal edges are longer yet does not exist in the original code
+        s[vertical_end:] += gamma
+        # Here there is a difference between the official MATLAB code and the paper
+        # on the edge penalty. We directly implement what the official code does.
 
         # Normalize differences
         s = self.normalize(s)
 
         # Gaussian weighting function
         s = -(
-            (np.exp(-beta * s, dtype="float64")) + 1.0e-6
-        )  # --> This epsilon changes results drastically default: 1.e-6
+            (np.exp(-beta * s, dtype="float64")) + 1e-5
+        )  # --> This epsilon changes results drastically default: 10e-6
+        # Please notice that it is not 1e-6, it is 10e-6 which is actually different.
 
         # Create Laplacian, diagonal missing
         lap = csc_matrix((s, (i, j)))
@@ -256,7 +280,14 @@ class UltrasoundConfidenceMap:
         return lap
 
     def _solve_linear_system(self, lap, rhs):
-        x = spsolve(lap, rhs)
+
+        if self.use_cg:
+            lap_sparse = lap.tocsr()
+            ml = ruge_stuben_solver(lap_sparse, coarse_solver="pinv")
+            m = ml.aspreconditioner(cycle="V")
+            x, _ = cg(lap, rhs, tol=self.cg_tol, maxiter=self.cg_maxiter, M=m)
+        else:
+            x = spsolve(lap, rhs)
 
         return x
 

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/losses/dice.py

@@ -811,7 +811,7 @@ class DiceFocalLoss(_Loss):
     The details of Focal Loss is shown in ``monai.losses.FocalLoss``.
 
     ``gamma`` and ``lambda_focal`` are only used for the focal loss.
-    ``include_background``, ``weight`` and ``reduction`` are used for both losses
+    ``include_background``, ``weight``, ``reduction``, and ``alpha`` are used for both losses,
     and other parameters are only used for dice loss.
 
     """
@@ -837,6 +837,7 @@ class DiceFocalLoss(_Loss):
         weight: Sequence[float] | float | int | torch.Tensor | None = None,
         lambda_dice: float = 1.0,
         lambda_focal: float = 1.0,
+        alpha: float | None = None,
     ) -> None:
         """
         Args:
@@ -871,7 +872,8 @@ class DiceFocalLoss(_Loss):
                 Defaults to 1.0.
             lambda_focal: the trade-off weight value for focal loss. The value should be no less than 0.0.
                 Defaults to 1.0.
-
+            alpha: value of the alpha in the definition of the alpha-balanced Focal loss. The value should be in
+                [0, 1]. Defaults to None.
         """
         super().__init__()
         weight = focal_weight if focal_weight is not None else weight
@@ -890,7 +892,12 @@ class DiceFocalLoss(_Loss):
             weight=weight,
         )
         self.focal = FocalLoss(
-            include_background=include_background, to_onehot_y=False, gamma=gamma, weight=weight, reduction=reduction
+            include_background=include_background,
+            to_onehot_y=False,
+            gamma=gamma,
+            weight=weight,
+            alpha=alpha,
+            reduction=reduction,
         )
         if lambda_dice < 0.0:
             raise ValueError("lambda_dice should be no less than 0.0.")

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/metrics/utils.py

@@ -35,9 +35,9 @@ from monai.utils import (
     optional_import,
 )
 
-binary_erosion, _ = optional_import("scipy.ndimage.morphology", name="binary_erosion")
-distance_transform_edt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_edt")
-distance_transform_cdt, _ = optional_import("scipy.ndimage.morphology", name="distance_transform_cdt")
+binary_erosion, _ = optional_import("scipy.ndimage", name="binary_erosion")
+distance_transform_edt, _ = optional_import("scipy.ndimage", name="distance_transform_edt")
+distance_transform_cdt, _ = optional_import("scipy.ndimage", name="distance_transform_cdt")
 
 __all__ = [
     "ignore_background",

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/transforms/intensity/array.py

@@ -2789,6 +2789,9 @@ class UltrasoundConfidenceMapTransform(Transform):
     It generates a confidence map by setting source and sink points in the image and computing the probability
     for random walks to reach the source for each pixel.
 
+    The official code is available at:
+    https://campar.in.tum.de/Main/AthanasiosKaramalisCode
+
     Args:
         alpha (float, optional): Alpha parameter. Defaults to 2.0.
         beta (float, optional): Beta parameter. Defaults to 90.0.
@@ -2796,14 +2799,32 @@ class UltrasoundConfidenceMapTransform(Transform):
         mode (str, optional): 'RF' or 'B' mode data. Defaults to 'B'.
         sink_mode (str, optional): Sink mode. Defaults to 'all'. If 'mask' is selected, a mask must be when
             calling the transform. Can be one of 'all', 'mid', 'min', 'mask'.
+        use_cg (bool, optional): Use Conjugate Gradient method for solving the linear system. Defaults to False.
+        cg_tol (float, optional): Tolerance for the Conjugate Gradient method. Defaults to 1e-6.
+            Will be used only if `use_cg` is True.
+        cg_maxiter (int, optional): Maximum number of iterations for the Conjugate Gradient method. Defaults to 200.
+            Will be used only if `use_cg` is True.
     """
 
-    def __init__(self, alpha: float = 2.0, beta: float = 90.0, gamma: float = 0.05, mode="B", sink_mode="all") -> None:
+    def __init__(
+        self,
+        alpha: float = 2.0,
+        beta: float = 90.0,
+        gamma: float = 0.05,
+        mode="B",
+        sink_mode="all",
+        use_cg=False,
+        cg_tol: float = 1.0e-6,
+        cg_maxiter: int = 200,
+    ):
         self.alpha = alpha
         self.beta = beta
         self.gamma = gamma
         self.mode = mode
         self.sink_mode = sink_mode
+        self.use_cg = use_cg
+        self.cg_tol = cg_tol
+        self.cg_maxiter = cg_maxiter
 
         if self.mode not in ["B", "RF"]:
             raise ValueError(f"Unknown mode: {self.mode}. Supported modes are 'B' and 'RF'.")
@@ -2813,7 +2834,9 @@ class UltrasoundConfidenceMapTransform(Transform):
                 f"Unknown sink mode: {self.sink_mode}. Supported modes are 'all', 'mid', 'min' and 'mask'."
             )
 
-        self._compute_conf_map = UltrasoundConfidenceMap(self.alpha, self.beta, self.gamma, self.mode, self.sink_mode)
+        self._compute_conf_map = UltrasoundConfidenceMap(
+            self.alpha, self.beta, self.gamma, self.mode, self.sink_mode, self.use_cg, self.cg_tol, self.cg_maxiter
+        )
 
     def __call__(self, img: NdarrayOrTensor, mask: NdarrayOrTensor | None = None) -> NdarrayOrTensor:
         """Compute confidence map from an ultrasound image.

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/transforms/signal/array.py

@@ -28,7 +28,7 @@ from monai.utils import optional_import
 from monai.utils.enums import TransformBackends
 from monai.utils.type_conversion import convert_data_type, convert_to_tensor
 
-shift, has_shift = optional_import("scipy.ndimage.interpolation", name="shift")
+shift, has_shift = optional_import("scipy.ndimage", name="shift")
 iirnotch, has_iirnotch = optional_import("scipy.signal", name="iirnotch")
 with warnings.catch_warnings():
     warnings.simplefilter("ignore", UserWarning)  # project-monai/monai#5204

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai/utils/module.py

@@ -13,6 +13,7 @@ from __future__ import annotations
 
 import enum
 import functools
+import importlib.util
 import os
 import pdb
 import re
@@ -208,9 +209,11 @@ def load_submodules(
     ):
         if (is_pkg or load_all) and name not in sys.modules and match(exclude_pattern, name) is None:
             try:
-                mod = import_module(name)
-                importer.find_spec(name).loader.load_module(name)  # type: ignore
-                submodules.append(mod)
+                mod_spec = importer.find_spec(name)  # type: ignore
+                if mod_spec and mod_spec.loader:
+                    mod = importlib.util.module_from_spec(mod_spec)
+                    mod_spec.loader.exec_module(mod)
+                    submodules.append(mod)
             except OptionalImportError:
                 pass  # could not import the optional deps., they are ignored
             except ImportError as e:

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426/monai_weekly.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: monai-weekly
-Version: 1.4.dev2424
+Version: 1.4.dev2426
 Summary: AI Toolkit for Healthcare Imaging
 Home-page: https://monai.io/
 Author: MONAI Consortium
@@ -70,6 +70,7 @@ Requires-Dist: zarr; extra == "all"
 Requires-Dist: lpips==0.1.4; extra == "all"
 Requires-Dist: nvidia-ml-py; extra == "all"
 Requires-Dist: huggingface_hub; extra == "all"
+Requires-Dist: pyamg>=5.0.0; extra == "all"
 Provides-Extra: nibabel
 Requires-Dist: nibabel; extra == "nibabel"
 Provides-Extra: ninja
@@ -145,6 +146,8 @@ Provides-Extra: pynvml
 Requires-Dist: nvidia-ml-py; extra == "pynvml"
 Provides-Extra: huggingface-hub
 Requires-Dist: huggingface_hub; extra == "huggingface-hub"
+Provides-Extra: pyamg
+Requires-Dist: pyamg>=5.0.0; extra == "pyamg"
 
 <p align="center">
   <img src="https://raw.githubusercontent.com/Project-MONAI/MONAI/dev/docs/images/MONAI-logo-color.png" width="50%" alt='project-monai'>

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/monai_weekly.egg-info/requires.txt

@@ -37,6 +37,7 @@ zarr
 lpips==0.1.4
 nvidia-ml-py
 huggingface_hub
+pyamg>=5.0.0
 
 [all:python_version <= "3.10"]
 transformers<4.41.0,>=4.36.0
@@ -120,6 +121,9 @@ pillow!=8.3.0
 [psutil]
 psutil
 
+[pyamg]
+pyamg>=5.0.0
+
 [pydicom]
 pydicom
 

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/pyproject.toml

@@ -38,7 +38,6 @@ exclude = "monai/bundle/__main__.py"
 
 [tool.ruff]
 line-length = 133
-lint.ignore-init-module-imports = true
 lint.ignore = ["F401", "E741"]
 
 [tool.pytype]

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/setup.cfg

@@ -81,6 +81,7 @@ all =
 	lpips==0.1.4
 	nvidia-ml-py
 	huggingface_hub
+	pyamg>=5.0.0
 nibabel = 
 	nibabel
 ninja = 
@@ -156,6 +157,8 @@ pynvml =
 	nvidia-ml-py
 huggingface_hub = 
 	huggingface_hub
+pyamg = 
+	pyamg>=5.0.0
 
 [flake8]
 select = B,C,E,F,N,P,T4,W,B9

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/tests/test_dice_focal_loss.py

@@ -91,6 +91,35 @@ class TestDiceFocalLoss(unittest.TestCase):
         test_input = torch.ones(2, 1, 8, 8)
         test_script_save(loss, test_input, test_input)
 
+    @parameterized.expand(
+        [
+            ("sum_None_0.5_0.25", "sum", None, 0.5, 0.25),
+            ("sum_weight_0.5_0.25", "sum", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("sum_weight_tuple_0.5_0.25", "sum", (3, 2.0, 1), 0.5, 0.25),
+            ("mean_None_0.5_0.25", "mean", None, 0.5, 0.25),
+            ("mean_weight_0.5_0.25", "mean", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("mean_weight_tuple_0.5_0.25", "mean", (3, 2.0, 1), 0.5, 0.25),
+            ("none_None_0.5_0.25", "none", None, 0.5, 0.25),
+            ("none_weight_0.5_0.25", "none", torch.tensor([1.0, 1.0, 2.0]), 0.5, 0.25),
+            ("none_weight_tuple_0.5_0.25", "none", (3, 2.0, 1), 0.5, 0.25),
+        ]
+    )
+    def test_with_alpha(self, name, reduction, weight, lambda_focal, alpha):
+        size = [3, 3, 5, 5]
+        label = torch.randint(low=0, high=2, size=size)
+        pred = torch.randn(size)
+
+        common_params = {"include_background": True, "to_onehot_y": False, "reduction": reduction, "weight": weight}
+
+        dice_focal = DiceFocalLoss(gamma=1.0, lambda_focal=lambda_focal, alpha=alpha, **common_params)
+        dice = DiceLoss(**common_params)
+        focal = FocalLoss(gamma=1.0, alpha=alpha, **common_params)
+
+        result = dice_focal(pred, label)
+        expected_val = dice(pred, label) + lambda_focal * focal(pred, label)
+
+        np.testing.assert_allclose(result, expected_val, err_msg=f"Failed on case: {name}")
+
 
 if __name__ == "__main__":
     unittest.main()

{monai_weekly-1.4.dev2424 → monai_weekly-1.4.dev2426}/tests/test_pad_collation.py

@@ -89,7 +89,7 @@ class TestPadCollation(unittest.TestCase):
 
     @parameterized.expand(TESTS)
     def test_pad_collation(self, t_type, collate_method, transform):
-        if t_type == dict:
+        if t_type is dict:
             dataset = CacheDataset(self.dict_data, transform, progress=False)
         else:
             dataset = _Dataset(self.list_data, self.list_labels, transform)
@@ -104,7 +104,7 @@ class TestPadCollation(unittest.TestCase):
         loader = DataLoader(dataset, batch_size=10, collate_fn=collate_method)
         # check collation in forward direction
         for data in loader:
-            if t_type == dict:
+            if t_type is dict:
                 shapes = []
                 decollated_data = decollate_batch(data)
                 for d in decollated_data:
@@ -113,7 +113,7 @@ class TestPadCollation(unittest.TestCase):
                     self.assertTrue(len(output["image"].applied_operations), len(dataset.transform.transforms))
                 self.assertTrue(len(set(shapes)) > 1)  # inverted shapes must be different because of random xforms
 
-        if t_type == dict:
+        if t_type is dict:
             batch_inverse = BatchInverseTransform(dataset.transform, loader)
             for data in loader:
                 output = batch_inverse(data)