Quantitative Digital Subtraction Angiography Measurement of Arterial Velocity at Low Radiation Dose Rates

Purpose: Quantitative digital subtraction angiography (qDSA) has been proposed to quantify blood velocity for monitoring treatment progress during blood flow altering interventions. The method requires high frame rate imaging [~ 30 frame per second (fps)] to capture temporal dynamics. This work investigates performance of qDSA in low radiation dose acquisitions to facilitate clinical translation.

Materials and methods: Velocity quantification accuracy was evaluated at five radiation dose rates in vitro and in vivo. Angiographic technique ranged from 30 fps digital subtraction angiography ( $29.3\pm 1.7\text{mGy}/\text{s}$ at the interventional reference point) down to a 30 fps protocol at 23% higher radiation dose per frame than fluoroscopy ( $1.1\pm 0.2\text{mGy}/\text{s}$ ). The in vitro setup consisted of a 3D-printed model of a swine hepatic arterial tree connected to a pulsatile displacement pump. Five different flow rates (3.5-8.8 mL/s) were investigated in vitro. Angiography-based fluid velocity measurements were compared across dose rates using ANOVA and Bland-Altman analysis. The experiment was then repeated in a swine study (n = 4).

Results: Radiation dose rate reductions for the lowest dose protocol were 99% and 96% for the phantom and swine study, respectively. No significant difference was found between angiography-based velocity measurements at different dose rates in vitro or in vivo. Bland-Altman analysis found little bias for all lower-dose protocols (range: [- 0.1, 0.1] cm/s), with the widest limits of agreement ([- 3.3, 3.5] cm/s) occurring at the lowest dose protocol.

Conclusions: This study demonstrates the feasibility of quantitative blood velocity measurements from angiographic images acquired at reduced radiation dose rates.