The logarithmic amplification of video signals and the availability of data in digital form make digital subtraction videoangiography a suitable tool for videodensitometric estimation of physiological quantities. A system for this purpose was implemented with a digital video image processor. However, it was found that the radiation scattering and veiling glare present in the image-intensified video must be removed to make meaningful quantitations. An algorithm to make such a correction was developed and is presented. With this correction, the videodensitometry system was calibrated with phantoms and used to measure the left ventricular ejection fraction of a canine heart.

Dual-energy subtraction imaging allows tissue and bone structures to be separated from each other and attenuating thicknesses measured. Potential applications include chest imaging, bone mineral measurement, angiography, and mammography. However, intrinsic to most x-ray detectors is the acceptance of scattered radiation as part of the image signal. Added to that is the veiling glare component when an image intensifier is used. Together, they result in erroneous transmission measurement and degrade the accuracy of energy subtraction processing. In this paper, the effects of scattered radiation and veiling glare on energy subtraction images are examined theoretically. A model is derived and used to compute the effects on the thickness signals, image contrast, and image noise as a function of the scatter glare to primary ratios. The ratios were measured on a point-by-point basis for a Rando chest phantom. For 96% of the image field studied, the thickness signals may be subject to an error ranging from 0 to -22.5 cm for tissue and 0 to 5.2 cm for bone. The image contrast in the tissue image may be reduced by a factor ranging from 1 to 59. The percentage of the uncanceled bone signals ranges from -52% to 52%. The contrast-to-noise ratio may be reduced by a factor ranging from 1 to 18.

Scattered radiation and veiling glare in digital subtraction angiography degrade contrast signals in a nonlinear and nonuniform way. This effect prohibits direct use of image data for accurate iodine measurement or energy subtraction imaging. Two techniques based on the use of scanning lead bars were proposed to measure the spatial distribution of scattered radiation and veiling glare in the unprocessed image data. With either technique, signals behind the lead bars were used to estimate the scatter-glare component and remove it from the image data. The scanning-lead-bar techniques can be used to reduce the scatter-glare component by a factor of up to 10. The scatter-glare correction leads to the recovery of degraded contrast signals and achieves reasonable linearity, uniformity, and consistency in the contrast measurement.

No abstract

No abstract