Abstract

Ventilation (V) and perfusion (Q) image data from 20 patients undergoing a routine clinical SPECT V/Q study were aligned via rigid coregistration using intrinsic image intensity values with the software package 'Qonsub'. Accuracy of the coregistration was quantified by an independent technique that used Tc-99m filled fiducial markers which had been adhered to the patient's skin prior to imaging. These were visible on both the V and Q scans, but were removed from the images for the coregistration step. The level of inter-marker displacement between scans was monitored to ensure it did not invalidate the assessment of coregistration accuracy. Once coregistered, results showed that for 65% of patients in the survey co-registration accuracy was better than 1 pixel (3.8 mm), 30% were co-registered with an accuracy of between 1 and 2 pixels and 5% were co-registered with an accuracy of between 2 and 3 pixels.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call