CODEWITHSUNDEEP
pythonnumpypixel
Kristan
Kristan

Reputation: 387

python dicom -- getting pixel value from dicom file

I was trying to get the pixel values of a dicom file in python using the dicom library.

But it returns only an array with zeros.

My code was like this:

import dicom

import numpy 

ds=pydicom.read_file("sample.dcm")

print(ds.pixel_array)

and the results is

 [[0 0 0 ..., 0 0 0]
 [0 0 0 ..., 0 0 0]
 [0 0 0 ..., 0 0 0]
 ..., 
 [0 0 0 ..., 0 0 0]
 [0 0 0 ..., 0 0 0]
 [0 0 0 ..., 0 0 0]]

Do you have any idea how to get the values of the pixels?

Upvotes: 2

Views: 4530

Answers (4)

Brian Mark Anderson
Brian Mark Anderson

Reputation: 546

One way you can visualize the image/dose that you're looking for is with the DicomRTTool (https://github.com/brianmanderson/Dicom_RT_and_Images_to_Mask)

You can install it with

pip install DicomRTTool

Then, use the code as follows to pull image/dose/structure information

from DicomRTTool.ReaderWriter import DicomReaderWriter, sitk
Dicom_path = r'.some_path_to_dicom'
Dicom_reader = DicomReaderWriter(get_dose_output=True)
Dicom_reader.walk_through_folders(Dicom_path) # This will parse through all DICOM present in the folder and subfolders

Dicom_reader.get_images()
Dicom_reader.get_dose()

image_sitk_handle = Dicom_reader.dicom_handle
image_numpy = sitk.GetArrayFromImage(image_sitk_handle)

dose_handle = Dicom_reader.dose_handle
dose_numpy = sitk.GetArrayFromImage(dose_handle)

You can then view the numpy array in whatever visualizer you prefer. I personally like Plot_And_Scroll_Images https://github.com/brianmanderson/Plot_And_Scroll_Images

from Plot_Scroll_Images import plot_scroll_Image
plot_scroll_Image(dose_numpy)

Hope this helps

Upvotes: 0

Aidos
Aidos

Reputation: 769

There should be values. If you use an IDE (i.g. Spyder) you may see it in the Variables Manager.

Upvotes: 0

jdg
jdg

Reputation: 2298

import dicom
import dicom_numpy


dataset = dicom.read_file('sample.dcm')
pixels, ijk_to_xyz = dicom_numpy.combine_slices([dataset])

Here pixels will contain the properly scaled pixel values, and ijk_to_xyz is a 4x4 affine matrix mapping the indices to the DICOM patient coordinate system.

This code snippet uses dicom-numpy, a higher-level abstraction built on top of pydicom.

It handles the most common use case when extracting pixel values from a set of DICOM files:

We use it on all our projects where we need to extract image data from DICOM files into a NumPy ndarray.

Upvotes: -1

darcymason
darcymason

Reputation: 1358

The code as written should work. It is possible that the beginning and end of the array are truly zeros.

It is usually more meaningful to look at something in the middle of the image.

E.g.:

midrow = ds.Rows // 2
midcol = ds.Columns // 2
print(ds.pixel_array[midrow-20:midrow+20, midcol-20:midcol+20])

If it is truly zeros everywhere, then that is either the true image or some kind of bug. Perhaps reading with a dicom viewer can help if that hasn't been tried.

Upvotes: 2

Related Questions