Pixel Spacing - DICOM

This is a discussion on Pixel Spacing - DICOM ; Hello, What is the difference between DICOM header Detector Element Spacing (0018,7022) and Imager Pixel Spacing (0018,1164)? The DICOM standard states that these two values may not be the same. What I would really like to know is what is ...

+ Reply to Thread
Results 1 to 4 of 4

Thread: Pixel Spacing

  1. Pixel Spacing

    Hello,

    What is the difference between DICOM header Detector Element Spacing
    (0018,7022) and Imager Pixel Spacing (0018,1164)? The DICOM standard
    states that these two values may not be the same. What I would really
    like to know is what is the spacing between each pixel in the DICOM
    image. Also, is attribute Detector Element Physical Size (0018,7020)
    most likely to be the same as Imager Pixel Spacing or Detector Element
    Spacing?

    Thank you for your help,

    Joel Feldman


  2. Imger versus Detector attributes, was Re: Pixel Spacing

    Hi Joel

    The short answer is that:

    - (0018,1164) describes a spacing equivalent to that which
    would be measured off a film in projection radiography

    - (0018,7022) does not describe the image pixels themselves,
    since detector elements may have been binned to produce
    pixels

    - (0018,7020) may be different from (0018,7022) since there
    may be non-sensitive material separating individual
    detectors (i.e. the size is smaller than the spacing
    between centers)

    Only (0018,1164) is relevant when measuring things; the
    detector-specific attributes are there to describe the
    acquisition.

    David

    PS. For ultrasound you need to use Region Calibration.

    feldman_joel@yahoo.com wrote:
    > Hello,
    >
    > What is the difference between DICOM header Detector Element Spacing
    > (0018,7022) and Imager Pixel Spacing (0018,1164)? The DICOM standard
    > states that these two values may not be the same. What I would really
    > like to know is what is the spacing between each pixel in the DICOM
    > image. Also, is attribute Detector Element Physical Size (0018,7020)
    > most likely to be the same as Imager Pixel Spacing or Detector Element
    > Spacing?
    >
    > Thank you for your help,
    >
    > Joel Feldman
    >


  3. Re: Imger versus Detector attributes, was Re: Pixel Spacing


    David Clunie wrote:

    Hi, David.
    Just one answer about Spacing :
    How must we use
    Pixel Aspect Ratio (0028,0034), for US images ?
    (I've some US images with a Pixel Aspect Ratio "4\3" that look OK if I
    don't deal with it, some other ones that are *not* OK at all.
    Is it an other DataElement to deal with?
    Thx a lot.
    Jean-Pierre Roux (from gdcm team)

    > Hi Joel
    >
    > The short answer is that:
    >
    > - (0018,1164) describes a spacing equivalent to that which
    > would be measured off a film in projection radiography
    >
    > - (0018,7022) does not describe the image pixels themselves,
    > since detector elements may have been binned to produce
    > pixels
    >
    > - (0018,7020) may be different from (0018,7022) since there
    > may be non-sensitive material separating individual
    > detectors (i.e. the size is smaller than the spacing
    > between centers)
    >
    > Only (0018,1164) is relevant when measuring things; the
    > detector-specific attributes are there to describe the
    > acquisition
    >
    > David
    >
    > PS. For ultrasound you need to use Region Calibration.
    >
    > feldman_joel@yahoo.com wrote:
    > > Hello,
    > >
    > > What is the difference between DICOM header Detector Element Spacing
    > > (0018,7022) and Imager Pixel Spacing (0018,1164)? The DICOM standard
    > > states that these two values may not be the same. What I would really
    > > like to know is what is the spacing between each pixel in the DICOM
    > > image. Also, is attribute Detector Element Physical Size (0018,7020)
    > > most likely to be the same as Imager Pixel Spacing or Detector Element
    > > Spacing?
    > >
    > > Thank you for your help,
    > >
    > > Joel Feldman
    > >



  4. Re: Imger versus Detector attributes, was Re: Pixel Spacing

    David,

    You mention:

    > Only (0018,1164) is relevant when measuring things; the
    > detector-specific attributes are there to describe the
    > acquisition.


    My question:

    Doesn't the Estimated Radiographic Maginification Factor (0018,1114)
    also play a role in creating accurate measurments when it is provided
    by the acquisition? Thus it is also relevant, when provided?

    I'm just checking up on a prior post from a while back and my
    understanding of these attributes being discussed.

    Homer


    David Clunie wrote:
    > Hi Joel
    >
    > The short answer is that:
    >
    > - (0018,1164) describes a spacing equivalent to that which
    > would be measured off a film in projection radiography
    >
    > - (0018,7022) does not describe the image pixels themselves,
    > since detector elements may have been binned to produce
    > pixels
    >
    > - (0018,7020) may be different from (0018,7022) since there
    > may be non-sensitive material separating individual
    > detectors (i.e. the size is smaller than the spacing
    > between centers)
    >
    > Only (0018,1164) is relevant when measuring things; the
    > detector-specific attributes are there to describe the
    > acquisition.
    >
    > David
    >
    > PS. For ultrasound you need to use Region Calibration.
    >
    > feldman_joel@yahoo.com wrote:
    > > Hello,
    > >
    > > What is the difference between DICOM header Detector Element Spacing
    > > (0018,7022) and Imager Pixel Spacing (0018,1164)? The DICOM standard
    > > states that these two values may not be the same. What I would really
    > > like to know is what is the spacing between each pixel in the DICOM
    > > image. Also, is attribute Detector Element Physical Size (0018,7020)
    > > most likely to be the same as Imager Pixel Spacing or Detector Element
    > > Spacing?
    > >
    > > Thank you for your help,
    > >
    > > Joel Feldman
    > >



+ Reply to Thread