Remove pinning highest versions of monai. (#560)

* Remove pinning highest versions of holoscan and monai.

This is done after having analyzed and found resolutions to issues
in example apps with monai v1.5.1, and after testing the fix to the
monai model inference configs that mix up RAS with LPS

Signed-off-by: M Q <mingmelvinq@nvidia.com>

* Still need to cap holoscan at v3.5.0 needing Python 3.9 for Doc build]

Signed-off-by: M Q <mingmelvinq@nvidia.com>

* Correct word used in constant name\

Signed-off-by: M Q <mingmelvinq@nvidia.com>

* Rerun notebook

Signed-off-by: M Q <mingmelvinq@nvidia.com>

---------

Signed-off-by: M Q <mingmelvinq@nvidia.com>

Author: MMelQin

monai/deploy/operators/dicom_series_to_volume_operator.py

@@ -39,6 +39,11 @@ class DICOMSeriesToVolumeOperator(Operator):
         image: Image object.
     """
 
+    # Use constants instead of enums in monai to avoid dependency at this level.
+    MONAI_UTIL_ENUMS_SPACEKEYS_LPS = "LPS"
+    MONAI_TRANSFORMS_SPATIAL_METADATA_NAME = "space"
+    METADATA_SPACE_LPS = {MONAI_TRANSFORMS_SPATIAL_METADATA_NAME: MONAI_UTIL_ENUMS_SPACEKEYS_LPS}
+
     def __init__(self, fragment: Fragment, *args, **kwargs):
         """Create an instance for a containing application object.
 
@@ -84,6 +89,18 @@ def convert_to_image(self, study_selected_series_list: List[StudySelectedSeries]
             metadata.update(self._get_instance_properties(study_selected_series.study))
             selection_metadata = {"selection_name": selection_name}
             metadata.update(selection_metadata)
+            # Add the metadata to specify LPS.
+            # Previously, this was set in ImageReader class, but moving it here allows other loaders
+            # to determine this value on its own, e.g. NIfTI loader but it does not set this
+            # resulting in the MONAI Orientation transform to default the labels to RAS.
+            # It is assumed that the ImageOrientationPatient will be set accordingly if the
+            # PatientPosition is other than HFS.
+            # NOTE: This value is properly parsed by MONAI Orientation transform from v1.5.1 onwards.
+            #       Some early MONAI model inference configs incorrectly specify orientation to RAS
+            #       due part to previous MONAI versions did not correctly parse this metadata from
+            #       the input MetaTensor and defaulting to RAS. Now with LPS properly set, the inference
+            #       configs then need to be updated to specify LPS, to achieve the same result.
+            metadata.update(self.METADATA_SPACE_LPS)
 
             voxel_data = self.generate_voxel_data(dicom_series)
             image = self.create_volumetric_image(voxel_data, metadata)

monai/deploy/operators/monai_bundle_inference_operator.py

@@ -614,7 +614,7 @@ def _init_config(self, config_names):
         )
 
     def _get_compose(self, obj_name, disallowed_prefixes):
-        """Gets a Compose object containing a sequence fo transforms from item `obj_name` in `self._parser`."""
+        """Gets a Compose object containing a sequence of transforms from item `obj_name` in `self._parser`."""
 
         if self._parser.get(obj_name) is not None:
             compose = self._parser.get_parsed_content(obj_name)
@@ -749,6 +749,7 @@ def compute(self, op_input, op_output, context):
                     raise ValueError("`meta_data` must be a dict.")
                 value = MetaTensor.ensure_torch_and_prune_meta(value, meta_data)
                 inputs[name] = value
+                logging.debug(f"Input MetaTensor metadata 'space': {value.meta.get('space', None)}")
                 # Named metadata dict not needed any more, as it is in the MetaTensor
 
             inputs = self.pre_process(inputs)
@@ -1062,7 +1063,8 @@ def _convert_from_image_dicom_source(self, img: Image) -> Tuple[np.ndarray, Dict
         # Use defines MetaKeys directly
         meta_dict[MetaKeys.ORIGINAL_AFFINE] = np.asarray(img_meta_dict.get("nifti_affine_transform", None))
         meta_dict[MetaKeys.AFFINE] = meta_dict[MetaKeys.ORIGINAL_AFFINE].copy()
-        meta_dict[MetaKeys.SPACE] = SpaceKeys.LPS  # not using SpaceKeys.RAS or affine_lps_to_ras
+        # Disabled setting the SPACE key below as it has changed to be set in the actual loader implementation.
+        # meta_dict[MetaKeys.SPACE] = SpaceKeys.LPS  # not using SpaceKeys.RAS or affine_lps_to_ras
 
         # Similarly the Image ndarray has dim order DHW, to be rearranged to WHD.
         # TODO: Need to revisit this once multi-channel image is supported and the Image class itself

monai/deploy/operators/monai_seg_inference_operator.py

@@ -90,7 +90,7 @@ def filter_sw_kwargs(**kwargs) -> Tuple[Dict[str, Any], Dict[str, Any]]:
             - Are not explicitly defined in the __init__ of this class
             - Are not explicitly used when calling sliding_window_inference
 
-        2. A dicionary of named parameters to pass to the base class __init__ of this class that:
+        2. A dictionary of named parameters to pass to the base class __init__ of this class that:
             - Are not used when calling sliding_window_inference
             - Can be successfully converted from Python --> Holoscan's C++ layer
 
@@ -582,7 +582,8 @@ def _get_meta_dict(self, img: Image) -> Dict:
         # Use define metadata kyes directly
         meta_dict[MetaKeys.ORIGINAL_AFFINE] = np.asarray(img_meta_dict.get("nifti_affine_transform", None))
         meta_dict[MetaKeys.AFFINE] = meta_dict[MetaKeys.ORIGINAL_AFFINE].copy()
-        meta_dict[MetaKeys.SPACE] = SpaceKeys.LPS  # not using SpaceKeys.RAS or affine_lps_to_ras
+        # Disabled setting the SPACE key below as it has changed to be set in the actual loader implementation.
+        # meta_dict[MetaKeys.SPACE] = SpaceKeys.LPS  # not using SpaceKeys.RAS or affine_lps_to_ras
         # The spatial shape, again, referring to ITKReader, it is the WHD
         meta_dict[MetaKeys.SPATIAL_SHAPE] = np.asarray(img.asnumpy().T.shape)
         # Well, no channel as the image data shape is forced to the the same as spatial shape