How to get the volume?

Each study is a physical slice through the LV. The dicom files have metadata, one item is a PixelSpacing [1,2]. The PixelSpacing provides the distance in mm between pixel centers for the row and column direction. Another bit of metadata is SliceThickness. This means that a pixel in a study represents a real volume where volume_pixel = (PixelSpacing^2 * SliceThickness). A potential approach for end-diastolic prediction: Identify all studies for a patient and check for full volume coverage/overlap (using dcm metadata??) Identify an image representing end-diastolic (biggest volume.. e.g. saxX…0001.dcm) in each study Identify pixels in the LV for each image Sum each pixel volume for pixels in the LV Adjust at the boundaries for volume elements that are not fully filled with LV (Full Disclosure: I did the Fourier…


Link to Full Article: How to get the volume?