Issues with WSI-properties dictionary and .tif Support #87
Description
Hi, hope you are doing well. I have encountered several issues with CellViT repo recently.
Describe the bug
WSI-properties dictionary is either ignored or leads to errors during pre-processing.
To Reproduce
Steps to reproduce the behavior:
- Command:
python ./preprocessing/patch_extraction/main_extraction.py --config ./example/preprocessing_example.yaml --wsi_properties "{\"slide_mpp\": \"0.5\", \"magnification\": \"20\"}"
Error Traceback:main_extraction.py: error: argument --wsi_properties: invalid dict value: '{"slide_mpp": "0.5", "magnification": "20"}'
Identical errors for --wsi_properties '{"slide_mpp": 0.5, "magnification": 20}' and --wsi_properties "dict(slide_mpp=0.5, magnification=20)"
- The alternative addition of it to config.yaml mentioned in preprocessing.md
wsi_properties:
- target_mpp: 0.5
- magnification: 20
Causes:
Traceback (most recent call last):
File "/cluster/CellViT/./preprocessing/patch_extraction/main_extraction.py", line 29, in <module>
configuration, logger = configuration_parser.get_config()
File "/cluster/CellViT/preprocessing/patch_extraction/src/cli.py", line 590, in get_config
yaml_config = PreProcessingYamlConfig(**yaml_config)
File "pydantic/main.py", line 342, in pydantic.main.BaseModel.__init__
pydantic.error_wrappers.ValidationError: 1 validation error for PreProcessingYamlConfig
wsi_properties
value is not a valid dict (type=type_error.dict)
The following options of config.yaml do not cause errors, but
target_mag: 20
target_mpp: 0.5
wsi_properties:
target_mpp: 0.5
magnification: 20
but are most likely ignored. Because when running preprocessing for a .tiff file that lacks metadata and using the config.yaml's from above:
OPENSLIDE_DEBUG=detection python ./preprocessing/patch_extraction/main_extraction.py --config ./example/preprocessing_example.yaml
I am getting:
2025-03-13 09:25:32,245 [INFO] - Using OpenSlide
2025-03-13 09:25:32,246 [INFO] - Data store directory: /cluster/CellViT/example/output/preprocessing
2025-03-13 09:25:32,246 [INFO] - Images found: 1
2025-03-13 09:25:32,246 [INFO] - Annotations found: 0
2025-03-13 09:25:32,246 [INFO] - Removing complete dataset! This may take a while.
2025-03-13 09:25:32,248 [INFO] - ****************************************************************************************************************************************************************************************************
2025-03-13 09:25:32,248 [INFO] - 1/1: MACEGEJ-1-1.tiff
2025-03-13 09:25:32,248 [INFO] - Computing patches for MACEGEJ-1-1.tiff
Openslide-Message: 09:25:32.248: mirax: Is a TIFF file
Openslide-Message: 09:25:32.249: hamamatsu-vms-vmu: Is a TIFF file
Openslide-Message: 09:25:32.249: hamamatsu-ndpi: No TIFF tag 65420
Openslide-Message: 09:25:32.249: sakura: Is a TIFF file
Openslide-Message: 09:25:32.249: trestle: No such value: directory 0, tag 305
Openslide-Message: 09:25:32.249: aperio: Not an Aperio slide
Openslide-Message: 09:25:32.249: leica: Not a Leica slide
Openslide-Message: 09:25:32.249: philips: No such value: directory 0, tag 305
Openslide-Message: 09:25:32.249: ventana: No such value: directory 0, tag 700
Openslide-Message: 09:25:32.253: mirax: Is a TIFF file
Openslide-Message: 09:25:32.253: hamamatsu-vms-vmu: Is a TIFF file
Openslide-Message: 09:25:32.253: hamamatsu-ndpi: No TIFF tag 65420
Openslide-Message: 09:25:32.253: sakura: Is a TIFF file
Openslide-Message: 09:25:32.254: trestle: No such value: directory 0, tag 305
Openslide-Message: 09:25:32.254: aperio: Not an Aperio slide
Openslide-Message: 09:25:32.254: leica: Not a Leica slide
Openslide-Message: 09:25:32.254: philips: No such value: directory 0, tag 305
Openslide-Message: 09:25:32.254: ventana: No such value: directory 0, tag 700
Traceback (most recent call last):
File "/cluster/CellViT/./preprocessing/patch_extraction/main_extraction.py", line 33, in <module>
slide_processor.sample_patches_dataset()
File "/cluster/CellViT/preprocessing/patch_extraction/src/patch_extraction.py", line 345, in sample_patches_dataset
) = self._prepare_wsi(wsi_file)
File "/cluster/CellViT/preprocessing/patch_extraction/src/patch_extraction.py", line 601, in _prepare_wsi
raise NotImplementedError(
NotImplementedError: MPP must be defined either by metadata or by config file!
Additional context
Actually, I started with a .tif file, but it throws an UnsupportedFromatError, even though there is a .tif option in wsi_extension:
2025-03-13 09:26:09,185 [INFO] - Using OpenSlide
2025-03-13 09:26:09,185 [INFO] - Data store directory: /cluster/CellViT/example/output/preprocessing
2025-03-13 09:26:09,186 [INFO] - Images found: 1
2025-03-13 09:26:09,186 [INFO] - Annotations found: 0
2025-03-13 09:26:09,186 [INFO] - Removing complete dataset! This may take a while.
2025-03-13 09:26:09,187 [INFO] - ****************************************************************************************************************************************************************************************************
2025-03-13 09:26:09,188 [INFO] - 1/1: MACEGEJ-1-1.tif
2025-03-13 09:26:09,188 [INFO] - Computing patches for MACEGEJ-1-1.tif
Openslide-Message: 09:26:09.287: mirax: Is a TIFF file
Openslide-Message: 09:26:09.287: hamamatsu-vms-vmu: Is a TIFF file
Openslide-Message: 09:26:09.287: hamamatsu-ndpi: No TIFF tag 65420
Openslide-Message: 09:26:09.287: sakura: Is a TIFF file
Openslide-Message: 09:26:09.288: trestle: Not a Trestle slide
Openslide-Message: 09:26:09.288: aperio: TIFF is not tiled
Openslide-Message: 09:26:09.288: leica: TIFF is not tiled
Openslide-Message: 09:26:09.288: philips: Not a Philips slide
Openslide-Message: 09:26:09.288: ventana: No such value: directory 0, tag 700
Openslide-Message: 09:26:09.288: generic-tiff: TIFF is not tiled
Traceback (most recent call last):
File "/cluster/CellViT/./preprocessing/patch_extraction/main_extraction.py", line 33, in <module>
slide_processor.sample_patches_dataset()
File "/cluster/CellViT/preprocessing/patch_extraction/src/patch_extraction.py", line 345, in sample_patches_dataset
) = self._prepare_wsi(wsi_file)
File "/cluster/CellViT/preprocessing/patch_extraction/src/patch_extraction.py", line 591, in _prepare_wsi
slide = OpenSlide(str(wsi_file))
File "/cluster/users/gmanaiev/conda/envs/cellvit/lib/python3.9/site-packages/openslide/__init__.py", line 166, in __init__
self._osr = lowlevel.open(str(filename))
File "/cluster/users/gmanaiev/conda/envs/cellvit/lib/python3.9/site-packages/openslide/lowlevel.py", line 195, in _check_open
raise OpenSlideUnsupportedFormatError("Unsupported or missing image file")
openslide.lowlevel.OpenSlideUnsupportedFormatError: Unsupported or missing image file
Module versions
All the modules have the required versions, and in particular:
openslide 3.4.1 h7773abc_6 https://conda.anaconda.org/conda-forge
openslide-python 1.2.0 pypi_0 pypi
libtiff 4.5.0 h6adf6a1_2 https://conda.anaconda.org/conda-forge