Purpose: To propose a method for reducing the radiation dose in cerebral CT perfusion studies by using a variable scan schedule.
Materials and methods: Original images were obtained with a multi detector-row CT (MDCT) scanner using the following CT perfusion protocol: continuous scans of 1 sec/rotation x 60 sec, four 5-mm-thick contiguous slices. The original images were thinned-out using combinations of various numbers of former continuous images (10, 15, 20, 25, and 30), and the later skipped images with various scan intervals (2, 5, 10, 15 and 20 sec). The thinned-out images were interpolated by linear interpolation. In five patients with cerebrovascular disease, we generated functional images of cerebral blood flow (CBF), cerebral blood volume (CBV), and mean transit time (MTT) from both original and interpolated data. The correlation coefficients (CC) for these parameters between the original and interpolated images were evaluated.
Results: The CC decreased with dose reduction. To keep the correlation coefficients greater than 0.9, the estimated dose was reduced to 33.3% on CBF with a set of 10 continuous images and scan interval of 5 sec, to 20.0% on CBV with a set of 10 continuous images and scan interval of 20 sec, and to 58.3% on MTT with a set of 10 continuous images and scan interval of 2 sec.
Conclusion: The variable scan schedule method would be useful to reduce radiation dose while maintaining the accuracy of CT perfusion (CTP) parameters.