Which Image preprocessing method is the best for Histology images?

I am working on classification of breast cancer histology images (Dataset:ICIAR BACH challenge 2018). But I don't really know which method is the best way to enhance the features of H&E stained image. I've tried the MATLAB code for Contrast limited adaptive histogram equalization method for enhancing the features. One of the images is attached below. This was my code:

I = imread('image.tif');

J = adapthisteq(I,'clipLimit',0.02,'Distribution','rayleigh');

But I don't think it's the right method to use. Can someone please suggest another MATLAB code for preprocessing the H&E stained image for best result?

(FYI I am completely new in the field of image processing and machine learning. and even though this post is related to image processing, I am asking this question here in MATLAB Answers because I am looking for a MATLAB code for processing the image.)

Thanks in advance.

Answer 1

[Image Analyst]

H&E color segmentation is covered here:

https://www.mathworks.com/help/images/color-based-segmentation-using-k-means-clustering.html

[Samiha Ahmed]

Thank you. But I don't want to segment the image. I only want to enhance the features.

[Image Analyst]

Enhance is a pretty broad word. I'd suggest converting to LAB or HSV color space and then using adapthisteq on the L or V channel ONLY. Then convert back to RGB color space. Something like

labImage = rgb2lab(rgbImage);
lImage = labImage(:, :, 1);
aImage = labImage(:, :, 2);
bImage = labImage(:, :, 3);
% Enahnce
limage = adapthisteq(lImage,'clipLimit',0.02,'Distribution','rayleigh');
labImage = cat(3, lImage, aImage, bImage);
rgbImageEnhanced = lab2rgb(labImage);

[Samiha Ahmed]

Thank you so much. I'll try this out.