how can I calculate the pixel value in DICOM image, if I consider each pixel a cube?

I have 300 stacks of DICOM images of aorta. I meshed the aorta, and I want to know the value of pixels which are within each element of aorta. what should I do?

[William Rose]

@Poorya,

Please explain what your ultimate goal is. Are you trying to determine wall density, wall thickness, lumen area, ...? Please provide a sample dicom image stack (maybe 3 slices, so it's not too large). Please provide a file of aortic mesh coordinates to accompany your three-slice image, so that others know what format your data has. Then others may be able to assist you.

[Poorya]

it is a liitle hard to explain. I am learning image registration and this is one of the step of learning.

I have 3D mesh volume of the aorta. so I have the position of each nodes. Since the voluma came from the DICOM image, I want to know which nodes are within which elements. for now, this is my goal. the next step for me is to find out the value of those pixel within each element, which it is easy as long as I solve the first step.

Thanks William

[William Rose]

@Poorya, I'm afraid I cannot assist you, because I do not understand the problem sufficiently. Others on this web site know a lot more than me about DICOM images. I think they will be able to assist you. They will be more likely to help if you post with a small example of your problem, with the associated files.

Answer 1

[Image Analyst]

By "3D mesh volume" do you mean "volumetric image"? By "node" do you mean "voxel"? By "element" do you mean the aorta voxels of the image? You will need to "segment" the image to produe a mask where true=aorta and false = non-aorta. I can't say how to do that without seeing the image but you may want to look at published papers:

http://www.visionbib.com/bibliography/contentsdescribe.html#3-D%20Object%20Description%20and%20Computation%20Techniques,%20Surfaces,%20Deformable,%20View%20Generation,%20Video%20Conferencing

and

http://www.visionbib.com/bibliography/contentsmedical.html#Medical%20Applications,%20CAT,%20MRI,%20Ultrasound,%20Heart%20Models,%20Brain%20Models

and you might take a look at imregister

help imregister

 imregister - Intensity-based image registration
    This MATLAB function transforms the 2-D or 3-D grayscale image, moving,
    so that it is registered with the reference image, fixed.

    Syntax
      moving_reg = imregister(moving,fixed,transformType,optimizer,metric)
      [moving_reg,R_reg] = imregister(moving,Rmoving,fixed,Rfixed,transformType,optimizer,metric)
      ___ = imregister(___,Name,Value)

    Input Arguments
      moving - Image to be registered
        numeric matrix | 3-D numeric array
      Rmoving - Spatial referencing information associated with image to be
      registered
        imref2d object | imref3d object
      fixed - Reference image
        numeric matrix | 3-D numeric array
      Rfixed - Spatial referencing information associated with reference
      image
        imref2d object | imref3d object
      transformType - Geometric transformation to be applied to image to be
      registered
        "translation" | "rigid" | "similarity" | "affine"
      optimizer - Method for optimizing similarity metric
        RegularStepGradientDescent or OnePlusOneEvolutionary optimizer object
      metric - Image similarity metric to be optimized during registration
        MeanSquares or MattesMutualInformation metric object

    Name-Value Arguments
      DisplayOptimization - Verbose optimization flag
        false (default) | true
      InitialTransformation - Initial geometric transformation
        affinetform2d object | affinetform3d object
      PyramidLevels - Number of pyramid levels used during registration
      process
        3 (default) | positive integer

    Output Arguments
      moving_reg - Registered image
        numeric matrix | 3-D numeric array
      R_reg - Spatial referencing information associated with registered
      image
        imref2d object | imref3d object

    Examples
      Register Multimodal MRI Images with Optimizer

    See also Registration Estimator, imregconfig, imregcorr, imregtform,
      imwarp, imshowpair, imfuse, imregicp

    Introduced in Image Processing Toolbox in R2012a
    Documentation for imregister
       doc imregister

[Poorya]

thank you for your response. that was helpful.