search

Project-MONAI / monai-deploy

MONAI Deploy aims to become the de-facto standard for developing, packaging, testing, deploying and running medical AI applications in clinical production.
Apache License 2.0
98 stars 22 forks source link

e2e test scenarios: KeyError: 'nifti_affine_transform' in liver_seg #118

Closed jordimassaguerpla closed 1 year ago

jordimassaguerpla commented 1 year ago

Hi,

I set up a monai cluster following this instructions https://github.com/Project-MONAI/monai-deploy-workflow-manager/pull/666 . In order to test it, I am using this clinical workflow from the e2e test scenarios https://github.com/Project-MONAI/monai-deploy/blob/main/e2e-testing/test-scenarios/Clinical_Workflows/liver_seg.json.

I download the images from this Medical Decathlon, as explained in https://github.com/Project-MONAI/monai-deploy/tree/main/deploy/monai-deploy-express#running-a-monai-deploy-workflow .

I rename the image work/monai-deploy-workflow-manager/deploy/examples/monai-deploy-lite-liver-ct/000587.dcm to input-dicom and uploaded this to Orthanc.

Then, I send the study to the Dicom modality and I get this error in the liver segmentation container:

Going to initiate execution of operator DICOMDataLoaderOperator
Executing operator DICOMDataLoaderOperator (Process ID: 25, Operator ID: e9628084-9846-44aa-b89
a-efb844fe1855)
[2023-02-08 13:36:52,828] [WARNING] (root) - Ignored /var/monai/input/1.2.826.0.1.3680043.2.112
5.1.19616861412188316212577695277886020/1.2.826.0.1.3680043.2.1125.1.34918616334750294149839565
085991567/1.2.826.0.1.3680043.2.1125.1.6173393152365463986791105716239686.dcm.json, reason bein
g: File is missing DICOM File Meta Information header or the 'DICM' prefix is missing from the
header. Use force=True to force reading.
Done performing execution of operator DICOMDataLoaderOperator

Going to initiate execution of operator DICOMSeriesSelectorOperator
Executing operator DICOMSeriesSelectorOperator (Process ID: 25, Operator ID: f6def06a-5cc6-4e0c
-9eaf-2c44340db6d9)
[2023-02-08 13:36:52,831] [INFO] (root) - Finding series for Selection named: CT Series
[2023-02-08 13:36:52,831] [INFO] (root) - Searching study, : 1.2.826.0.1.3680043.2.1125.1.19616
861412188316212577695277886020
  # of series: 1
[2023-02-08 13:36:52,831] [INFO] (root) - Working on series, instance UID: 1.2.826.0.1.3680043.
2.1125.1.34918616334750294149839565085991567
[2023-02-08 13:36:52,831] [INFO] (root) - On attribute: 'Modality' to match value: '(?i)CT'
[2023-02-08 13:36:52,831] [INFO] (root) -     Series attribute Modality value: CT
[2023-02-08 13:36:52,831] [INFO] (root) - Series attribute string value did not match. Try regE
x.
[2023-02-08 13:36:52,831] [INFO] (root) - On attribute: 'ImageType' to match value: '['PRIMARY'
, 'ORIGINAL']'
[2023-02-08 13:36:52,831] [INFO] (root) -     Series attribute ImageType value: None
[2023-02-08 13:36:52,831] [INFO] (root) - On attribute: 'PhotometricInterpretation' to match va
lue: 'MONOCHROME2'
[2023-02-08 13:36:52,831] [INFO] (root) -     Series attribute PhotometricInterpretation value:
 None
[2023-02-08 13:36:52,831] [INFO] (root) - Selected Series, UID: 1.2.826.0.1.3680043.2.1125.1.34
918616334750294149839565085991567
Done performing execution of operator DICOMSeriesSelectorOperator

Going to initiate execution of operator DICOMSeriesToVolumeOperator
Executing operator DICOMSeriesToVolumeOperator (Process ID: 25, Operator ID: 5036c2da-fcc6-4f9a
-abf8-a59b70c28dd6)
Done performing execution of operator DICOMSeriesToVolumeOperator

Going to initiate execution of operator LiverTumorSegOperator
Executing operator LiverTumorSegOperator (Process ID: 25, Operator ID: 0b577306-331d-48f7-a974-
9b462b54a155)
Operator output folder path: /var/monai/operators/0b577306-331d-48f7-a974-9b462b54a155/0/output
/saved_images_folder
Converted Image object metadata:
SeriesInstanceUID: 1.2.826.0.1.3680043.2.1125.1.34918616334750294149839565085991567, type <clas
s 'str'>
SeriesDate: 20201014, type <class 'str'>
SeriesTime: 162720, type <class 'str'>
Modality: CT, type <class 'str'>
SeriesDescription: CT series for liver tumor from nii 014, type <class 'str'>
PatientPosition: HFS, type <class 'str'>
SeriesNumber: 1, type <class 'int'>
row_pixel_spacing: 0.685546875, type <class 'float'>
col_pixel_spacing: 0.685546875, type <class 'float'>
depth_pixel_spacing: 1.0, type <class 'float'>
row_direction_cosine: [1.0, 0.0, 0.0], type <class 'list'>
col_direction_cosine: [0.0, 1.0, 0.0], type <class 'list'>
StudyInstanceUID: 1.2.826.0.1.3680043.2.1125.1.19616861412188316212577695277886020, type <class
 'str'>
StudyID: SLICER10001, type <class 'str'>
StudyDate: 20201014, type <class 'str'>
StudyTime: 162720, type <class 'str'>
StudyDescription: CT Study for liver_14, type <class 'str'>
AccessionNumber: 1, type <class 'str'>
selection_name: CT Series, type <class 'str'>
2023-02-08 13:36:52,872 INFO image_writer.py:193 - writing: /var/monai/operators/0b577306-331d-
48f7-a974-9b462b54a155/0/output/saved_images_folder/1.2.826.0.1.3680043.2.1125.1/1.2.826.0.1.36
80043.2.1125.1.nii.gz
2023-02-08 13:36:58,139 INFO image_writer.py:193 - writing: /var/monai/operators/0b577306-331d-
48f7-a974-9b462b54a155/0/output/saved_images_folder/1.2.826.0.1.3680043.2.1125.1/1.2.826.0.1.36
80043.2.1125.1_seg.nii.gz
Output Seg image numpy array shaped: (1, 512, 512)
Output Seg image pixel max value: 0
Done performing execution of operator LiverTumorSegOperator

Going to initiate execution of operator DICOMSegmentationWriterOperator
Executing operator DICOMSegmentationWriterOperator (Process ID: 25, Operator ID: e616ac55-6c84-
4e00-a535-f26243e0b670)
[2023-02-08 13:36:58,481] [WARNING] (highdicom.seg.sop) - Encoding an empty segmentation with "
omit_empty_frames" set to True. Reverting to encoding all frames since omitting all frames is n
ot possible.
[2023-02-08 13:36:58,482] [INFO] (highdicom.seg.sop) - add plane #0 for segment #1
[2023-02-08 13:36:58,483] [INFO] (highdicom.seg.sop) - add plane #0 for segment #2
[2023-02-08 13:36:58,484] [INFO] (highdicom.base) - copy Image-related attributes from dataset
"1.2.826.0.1.3680043.2.1125.1.6173393152365463986791105716239686"
[2023-02-08 13:36:58,484] [INFO] (highdicom.base) - copy attributes of module "Specimen"
[2023-02-08 13:36:58,484] [INFO] (highdicom.base) - copy Patient-related attributes from datase
t "1.2.826.0.1.3680043.2.1125.1.6173393152365463986791105716239686"
[2023-02-08 13:36:58,484] [INFO] (highdicom.base) - copy attributes of module "Patient"
[2023-02-08 13:36:58,485] [INFO] (highdicom.base) - copy attributes of module "Clinical Trial S
ubject"
[2023-02-08 13:36:58,485] [INFO] (highdicom.base) - copy Study-related attributes from dataset
"1.2.826.0.1.3680043.2.1125.1.6173393152365463986791105716239686"
[2023-02-08 13:36:58,485] [INFO] (highdicom.base) - copy attributes of module "General Study"
[2023-02-08 13:36:58,485] [INFO] (highdicom.base) - copy attributes of module "Patient Study"
[2023-02-08 13:36:58,485] [INFO] (highdicom.base) - copy attributes of module "Clinical Trial S
tudy"
Done performing execution of operator DICOMSegmentationWriterOperator

Going to initiate execution of operator STLConversionOperator
Executing operator STLConversionOperator (Process ID: 25, Operator ID: 42c57850-90f7-4eef-8b75-5a00f43a97b9)
[2023-02-08 13:36:58,495] [INFO] (monai.deploy.operators.stl_conversion_operator.STLConversionOperator) - Output will be saved in file /var/monai/output/stl/liver_seg.stl.
Traceback (most recent call last):
  File "/opt/monai/app/app.py", line 142, in <module>
    app_instance.run()
  File "/opt/monai/app/app.py", line 62, in run
    super().run(*args, **kwargs)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/core/application.py", line 429, in run
    executor_obj.run()
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/core/executors/single_process_executor.py", line 125, in run
    op.compute(op_exec_context.input_context, op_exec_context.output_context, op_exec_context)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 95, in compute
    stl_bytes = self._convert(input_image, _output_file)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 121, in _convert
    return self._converter.convert(
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 165, in convert
    s_image = self.SpatialImage(image)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 297, in __init__
    self._read_from_in_mem_image(self._image)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 394, in _read_from_in_mem_image
    img_array, affine, original_affine, shape, spacing, itk_image = self._load_data(image)
  File "/root/.local/lib/python3.8/site-packages/monai/deploy/operators/stl_conversion_operator.py", line 366, in _load_data
    original_affine = img_meta_dict["nifti_affine_transform"]
KeyError: 'nifti_affine_transform'

What am I doing wrong?

Thanks

jordimassaguerpla commented 1 year ago

I just pushed all the dcm files in that zip instead of just on dcm file and I did not get the error!! Sorry, I am new to all this. Do I need all the 588 dcm files?