Add Clara Viz operator.

Signed-off-by: Andreas Heumann <aheumann@nvidia.com>

Author: AndreasHeumann

monai/deploy/operators/__init__.py

@@ -1,4 +1,4 @@
-# Copyright 2021 MONAI Consortium
+# Copyright 2021-2022 MONAI Consortium
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
 # You may obtain a copy of the License at
@@ -12,6 +12,7 @@
 .. autosummary::
     :toctree: _autosummary
 
+    ClaraVizOperator
     DICOMDataLoaderOperator
     DICOMSegmentationWriterOperator
     DICOMSeriesSelectorOperator
@@ -22,6 +23,7 @@
     PublisherOperator
 """
 
+from .clara_viz_operator import ClaraVizOperator
 from .dicom_data_loader_operator import DICOMDataLoaderOperator
 from .dicom_seg_writer_operator import DICOMSegmentationWriterOperator
 from .dicom_series_selector_operator import DICOMSeriesSelectorOperator

monai/deploy/operators/clara_viz_operator.py

@@ -0,0 +1,134 @@
+# Copyright 2022 MONAI Consortium
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#     http://www.apache.org/licenses/LICENSE-2.0
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import numpy as np
+
+import monai.deploy.core as md
+from monai.deploy.core import ExecutionContext, Image, InputContext, IOType, Operator, OutputContext
+from monai.deploy.utils.importutil import optional_import
+
+DataDefinition, _ = optional_import("clara.viz.core", name="DataDefinition")
+Widget, _ = optional_import("clara.viz.widgets", name="Widget")
+display, _ = optional_import("IPython.display", name="display")
+interactive, _ = optional_import("ipywidgets", name="interactive")
+Dropdown, _ = optional_import("ipywidgets", name="Dropdown")
+Box, _ = optional_import("ipywidgets", name="Box")
+VBox, _ = optional_import("ipywidgets", name="VBox")
+
+
+@md.input("image", Image, IOType.IN_MEMORY)
+@md.input("seg_image", Image, IOType.IN_MEMORY)
+@md.env(pip_packages=["clara.viz.core", "clara.viz.widgets", "IPython"])
+class ClaraVizOperator(Operator):
+    """
+    This operator uses Clara Viz to provide interactive view of a 3D volume including segmentation mask.
+    """
+
+    def __init__(self):
+        """Constructor of the operator."""
+        super().__init__()
+
+    @staticmethod
+    def _build_array(image, order):
+        numpy_array = image.asnumpy()
+
+        array = DataDefinition.Array(array=numpy_array, order=order)
+        array.element_size = [1.0]
+        array.element_size.append(image.metadata().get("col_pixel_spacing", 1.0))
+        array.element_size.append(image.metadata().get("row_pixel_spacing", 1.0))
+        array.element_size.append(image.metadata().get("depth_pixel_spacing", 1.0))
+
+        # the renderer is expecting data in RIP order (Right Inferior Posterior) which results in
+        # this matrix
+        target_affine_transform = [
+            [-1.0, 0.0, 0.0, 0.0],
+            [0.0, 0.0, 1.0, 0.0],
+            [0.0, -1.0, 0.0, 0.0],
+            [0.0, 0.0, 0.0, 1.0],
+        ]
+
+        dicom_affine_transform = image.metadata().get("dicom_affine_transform", np.identity(4))
+
+        affine_transform = np.matmul(target_affine_transform, dicom_affine_transform)
+
+        array.permute_axes = [
+            0,
+            max(range(3), key=lambda k: abs(affine_transform[0][k])) + 1,
+            max(range(3), key=lambda k: abs(affine_transform[1][k])) + 1,
+            max(range(3), key=lambda k: abs(affine_transform[2][k])) + 1,
+        ]
+
+        array.flip_axes = [
+            False,
+            affine_transform[0][array.permute_axes[1] - 1] < 0.0,
+            affine_transform[1][array.permute_axes[2] - 1] < 0.0,
+            affine_transform[2][array.permute_axes[3] - 1] < 0.0,
+        ]
+
+        return array
+
+    def compute(self, op_input: InputContext, op_output: OutputContext, context: ExecutionContext):
+        """Displays the input image and segmentation mask
+
+        Args:
+            op_input (InputContext): An input context for the operator.
+            op_output (OutputContext): An output context for the operator.
+            context (ExecutionContext): An execution context for the operator.
+        """
+        input_image = op_input.get("image")
+        if not input_image:
+            raise ValueError("Input image is not found.")
+        input_seg_image = op_input.get("seg_image")
+        if not input_seg_image:
+            raise ValueError("Input segmentation image is not found.")
+
+        # build the data definition
+        data_definition = DataDefinition()
+
+        data_definition.arrays.append(self._build_array(input_image, "DXYZ"))
+
+        data_definition.arrays.append(self._build_array(input_seg_image, "MXYZ"))
+
+        widget = Widget()
+        widget.select_data_definition(data_definition)
+        # default view mode is 'CINEMATIC' switch to 'SLICE_SEGMENTATION' since we have no transfer functions defined
+        widget.settings["Views"][0]["mode"] = "SLICE_SEGMENTATION"
+        widget.settings["Views"][0]["cameraName"] = "Top"
+        widget.set_settings()
+
+        # add controls
+        def set_view_mode(view_mode):
+            widget.settings["Views"][0]["mode"] = view_mode
+            if view_mode == "CINEMATIC":
+                widget.settings["Views"][0]["cameraName"] = "Perspective"
+            elif widget.settings["Views"][0]["cameraName"] == "Perspective":
+                widget.settings["Views"][0]["cameraName"] = "Top"
+            widget.set_settings()
+
+        widget_view_mode = interactive(
+            set_view_mode,
+            view_mode=Dropdown(
+                options=[("Cinematic", "CINEMATIC"), ("Slice", "SLICE"), ("Slice Segmentation", "SLICE_SEGMENTATION")],
+                value="SLICE_SEGMENTATION",
+                description="View mode",
+            ),
+        )
+
+        def set_camera(camera):
+            if widget.settings["Views"][0]["mode"] != "CINEMATIC":
+                widget.settings["Views"][0]["cameraName"] = camera
+                widget.set_settings()
+
+        widget_camera = interactive(
+            set_camera, camera=Dropdown(options=["Top", "Right", "Front"], value="Top", description="Camera")
+        )
+
+        display(Box([widget, VBox([widget_view_mode, widget_camera])]))