kidzik / som

Cell segmentation self-organizing maps

Sort nucleis in H&E slides according to their shapes.

Usage

For testing on the example slide from input directory:

1. Clone this repository
2. Install requirements.txt for running example analysis notebook
3. Run the segsom-example-notebook.ipynb notebook

For running analysis on your own image:

1. Segment the HE using https://hub.docker.com/r/kidzik/he_cell_seg (or if you have NVIDIA GPU, use a GPU-accelerated version https://hub.docker.com/r/kidzik/he_cell_seg_gpu)
2. Update file lecation and segmentation mask location in the notebook
3. Run the segsom-example-notebook.ipynb notebook

Credits

Code in this repository by Edwin Yuan and Magdalena Matusiak.

Cell segmentation scripts and the neural network by Korsuk Sirinukunwattana.

som's Issues

The example notebook doesn't work. Example two mat files may be corrupted.

I cloned this repository and installed requirement.txt. Then I run segsom-example-notebook.ipynb.
But this error occured.

`
somhandle = SOM()
somhandle.readtrainedSOM('som_models/SOM_v3_12_13/trainedvecs7x7.mat')
somhandle.readPCAcoef('som_models/SOM_v3_12_13/vectors.mat')

---

ValueError Traceback (most recent call last)
/var/folders/gj/mr_1z3xs3zsgccyc2rllwj240000gn/T/ipykernel_6620/4268023471.py in
1 somhandle = SOM()
----> 2 somhandle.readtrainedSOM('som_models/SOM_v3_12_13/trainedvecs7x7.mat')
3 somhandle.readPCAcoef('som_models/SOM_v3_12_13/vectors.mat')

~/som/modules/SOMclass2.py in readtrainedSOM(self, path)
73 # Input:
74 # - path - path of the trained SOM vectors
---> 75 self.SOMvecs = loadmat(path)['trainedvecs7x7'].T #(1000,49)
76
77 def readPCAcoef(self, path):

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/mio.py in loadmat(file_name, mdict, appendmat, **kwargs)
221 variable_names = kwargs.pop('variable_names', None)
222 with _open_file_context(file_name, appendmat) as f:
--> 223 MR, _ = mat_reader_factory(f, **kwargs)
224 matfile_dict = MR.get_variables(variable_names)
225

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/mio.py in mat_reader_factory(file_name, appendmat, **kwargs)
70 """
71 byte_stream, file_opened = _open_file(file_name, appendmat)
---> 72 mjv, mnv = get_matfile_version(byte_stream)
73 if mjv == 0:
74 return MatFile4Reader(byte_stream, **kwargs), file_opened

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/miobase.py in get_matfile_version(fileobj)
229 if maj_val in (1, 2):
230 return ret
--> 231 raise ValueError('Unknown mat file type, version %s, %s' % ret)
232
233

ValueError: Unknown mat file type, version 51, 55
`

And I run this code to check whether these mat files are corrupted.

`
import scipy.io
scipy.io.loadmat('som_models/SOM_v3_12_13/trainedvecs7x7.mat')

---

ValueError Traceback (most recent call last)
/var/folders/gj/mr_1z3xs3zsgccyc2rllwj240000gn/T/ipykernel_6620/3498248664.py in
1 import scipy.io
----> 2 scipy.io.loadmat('som_models/SOM_v3_12_13/trainedvecs7x7.mat')

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/mio.py in loadmat(file_name, mdict, appendmat, **kwargs)
221 3.14159265+3.14159265j])
222 """
--> 223 variable_names = kwargs.pop('variable_names', None)
224 with _open_file_context(file_name, appendmat) as f:
225 MR, _ = mat_reader_factory(f, **kwargs)

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/mio.py in mat_reader_factory(file_name, appendmat, **kwargs)
70 Whether the file was opened by this routine.
71
---> 72 """
73 byte_stream, file_opened = _open_file(file_name, appendmat)
74 mjv, mnv = get_matfile_version(byte_stream)

~/opt/anaconda3/envs/som/lib/python3.8/site-packages/scipy/io/matlab/miobase.py in get_matfile_version(fileobj)
229 if maj_val in (1, 2):
230 return ret
--> 231 raise ValueError('Unknown mat file type, version %s, %s' % ret)
232
233

ValueError: Unknown mat file type, version 51, 55
`

Thank you.