IHC Image color deconvolution

[Koundinya Desiraju]

In this tutorial, http://scikit-image.org/docs/dev/auto_examples/color_exposure/plot_ihc_color_separation.html#sphx-glr-auto-examples-color-exposure-plot-ihc-color-separation-py

After color deconvolution the haematoxylin and DAB channels were plotted in grayscale but I want to show them in respective colors. For example, I want to show DAB channel in brown. Can anyone kindly post How to do that?

Thanks,

Koundinya

[Juan Nunez-Iglesias]

Hi Koundinya,

It looks like this is not a trivial question, because the conversion between the two spaces is nonlinear. See the code here:

https://github.com/scikit-image/scikit-image/blob/master/skimage/color/colorconv.py#L1437

I originally thought that it would be possible to convert each channel separately, but the `exp` call makes this either difficult or impossible (I haven't yet decided which, but I wasn't able to get it to work in a short amount of time).

Here's my initial attempt, which was a complete failure, for the record:

In [1]: from skimage import data

In [2]: from skimage.color import rgb2hed

In [3]: from skimage.color import hed2rgb

In [4]: import matplotlib.pyplot as plt

In [5]: ihc_rgb = data.immunohistochemistry()

In [6]: ihc_hed = rgb2hed(ihc_rgb)

In [8]: import numpy as np

In [9]: h, e, d = np.transpose(ihc_hed, (2, 0, 1))

In [10]: empty = np.zeros_like(h)

In [11]: h_rgb = hed2rgb(np.dstack((h, empty, empty)))

In [12]: e_rgb = hed2rgb(np.dstack((empty, e, empty)))

In [13]: d_rgb = hed2rgb(np.dstack((empty, empty, d)))

In [14]: fig, axes = plt.subplots(2, 2, figsize=(6, 7), sharex=True,

    ...:                          sharey=True,

    ...:                          subplot_kw={'adjustable': 'box-forced'})

In [15]: axes[0, 0].imshow(ihc_rgb); axes[0, 0].set_title('Original image')

Out[15]: <matplotlib.text.Text at 0x121357438>

In [16]: axes[0, 1].imshow(h_rgb); axes[0, 1].set_title('Hematoxylin')

Out[16]: <matplotlib.text.Text at 0x121371c18>

In [17]: axes[1, 0].imshow(e_rgb); axes[1, 0].set_title('Eosin')

Out[17]: <matplotlib.text.Text at 0x1210684e0>

In [18]: axes[1, 1].imshow(d_rgb); axes[1, 1].set_title('DAB')

Out[18]: <matplotlib.text.Text at 0x1219315f8>

In [19]: for ax in axes.ravel():

    ...:     ax.axis('off')

Result:

Maybe someone else in the list has some ideas?

Juan.

--
You received this message because you are subscribed to the Google Groups "scikit-image" group.
To unsubscribe from this group and stop receiving emails from it, send an email to scikit-image...@googlegroups.com.
To post to this group, send email to scikit...@googlegroups.com.
To view this discussion on the web, visit https://groups.google.com/d/msgid/scikit-image/74882dde-6af5-4cb8-a02d-debd5e2e1678%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Cedric Espenel]

Hi Koundinya,

I'm not sure if that will answer your question but if you try:

from matplotlib.colors import LinearSegmentedColormap
cmap_Hema = LinearSegmentedColormap.from_list('mycmap', ['white', 'navy'])
cmap_DAB = LinearSegmentedColormap.from_list('mycmap', ['white', 
'saddlebrown'])
cmap_Eosin = LinearSegmentedColormap.from_list('mycmap', ['darkviolet', 
'white'])

You won't be able to get back to the original color but it will do a descent job: 

Sincerely,

Cedric

[Koundinya Desiraju]

Thank you Cedric. That helped a lot.

Regards,

Koundinya

--
You received this message because you are subscribed to a topic in the Google Groups "scikit-image" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/scikit-image/yumvSsa7irc/unsubscribe.
To unsubscribe from this group and all its topics, send an email to scikit-image...@googlegroups.com.

To post to this group, send email to scikit...@googlegroups.com.

To view this discussion on the web, visit https://groups.google.com/d/msgid/scikit-image/5394a2f0-98a1-4254-ac14-851e9c956bb7%40googlegroups.com.

[Stefan van der Walt]

On Tue, Aug 30, 2016, at 00:56, Koundinya Desiraju wrote:

Thank you Cedric. That helped a lot.

Would you like to add this to the gallery example, so that other users can do the same?

Stéfan

[Stefan van der Walt]

On Mon, Aug 22, 2016, at 22:40, Koundinya Desiraju wrote:

In this tutorial, http://scikit-image.org/docs/dev/auto_examples/color_exposure/plot_ihc_color_separation.html#sphx-glr-auto-examples-color-exposure-plot-ihc-color-separation-py

After color deconvolution the haematoxylin and DAB channels were plotted in grayscale but I want to show them in respective colors. For example, I want to show DAB channel in brown. Can anyone kindly post How to do that?

Cedric submitted a PR to add this to the examples gallery: https://github.com/scikit-image/scikit-image/pull/2279

Please review!

Thanks
Stéfan