Skip to content
chafey edited this page Mar 21, 2014 · 9 revisions

The image object describes the image that is to be rendered.

  • minPixelValue - the minimum stored pixel value in the image
  • maxPixelValue - the maximum stored pixel value in the image
  • slope - the rescale slope to convert stored pixel values to modality pixel values or 1 if not specified
  • intercept - the rescale intercept used to convert stored pixel values to modality values or 0 if not specified
  • windowCenter - the default windowCenter to apply to the image
  • windowWidth - the default windowWidth to apply to the image
  • rows - the number of rows in the image (height)
  • columns - the number of columns in the image (width)
  • color - true if pixel data is RGB, false if grayscale

NOTE: It would be nice to remove some of DICOM's complexity here if we can - for example, make all pixel data unsigned, MONOCHROME2 and set pixelPadding to the minimum pixel value (or lower). This could all be applied on by the ImageLoader plugin to make cornerstone simpler and faster. If we don't go this route, the following attributes could be added to expose this complexity

  • darkToBright - false if image is grayscale and Photometric Interpretation is MONOCHROME1, true otherwise
  • bitsAllocated - number of bits allocated per pixel.
  • bitsStored - number of bits stored per pixel
  • highBit - the highest bit stored
  • pixelRepresentation - 0 unsigned, 1 signed
  • pixelPadding - values not to display (always black)