how to separate the tumor portion from the image?

30 views (last 30 days)
I am detecting brain tumor b using median filter watershed segmentation. I converted the image to grey,remove the noise using median filter, adjusted the image,converted it to lack and white and then took the compliment. Then i applied the distance transform and use watershed segmentation. I want to know how to extract the tumor infected area from the image?? And also want to know how to calculate the area of the infected area.? Please tell me...
  2 Comments
Jhilam Mukherjee
Jhilam Mukherjee on 21 Oct 2015
if your image is a binary image apply regionprops
jasleen sha
jasleen sha on 22 Oct 2015
yes i am applying regionprops command but its not working for me... May be I'm applying in wrong way...

Sign in to comment.

Answers (1)

Image Analyst
Image Analyst on 21 Oct 2015
You said you "converted it to lack and white" so you've already segmented the tumor into a binary image, so now we just need to know what you mean by "extract the tumor". Do you mean like crop out, or mask the exterior area to black???
To compute area, just use regionprops on the binary image. See my Image Segmentation Tutorial for an example. http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862
I'm not sure why you did a distance transform and then watershed on that. What was the point in doing that? I mean, you already have the segmented, binary image so what are all these subsequent operations for?
  4 Comments
Shyme Bhate
Shyme Bhate on 31 Mar 2020
Hi, Image Analyst. I need little guidance, As I want to extract tumorius part from the brain MRI , I am extracting that part by split the images but how to know the exact coordinates of the image ?
Image Analyst
Image Analyst on 31 Mar 2020
You can ask regionprops() for PixelList and it will give you the coordinates of every pixel in the blob you're interested in. Start a new question if you still have trouble. Also, see attached demo.

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!