Room: Track 1
Purpose: interventional procedures require accurate intraprocedural measurement of cerebral vasculature structures. However, cardiac-induced motion within vessels and arteries introduce image distortion, reducing the accuracy of geometric measurements for implant and device sizing. Here, we focus on the carotid artery and propose the use of our novel dynamic imaging technique Adaptive CaRdiac cOne BEAm computed Tomography (ACROBEAT) to limit cardiac-induced image distortion and reduce imaging dose. ACROBEAT adapts the gantry velocity and projection rate of the imaging system in real-time with changes in a patient’s electrocardiogram (ECG).
Methods: designed and manufactured a carotid artery phantom from an 8mm diameter silicon rubber tubing cast into phytagel. An artery motion curve, derived from patient-measured ECG trace with average heart rate of 56 bpm, was programmed into the phantom, enabling replication of realistic patient anatomy and physiology. The phantom was imaged on a Siemens ARTIS pheno robotic angiography system in conjunction with a Siemens Test Automation Control System. Two methods were implemented: (1) a single gantry rotation with constant speed and acquisition rate which is the standard of care for intracranial procedures and (2) ACROBEAT. The width of the artery in the reconstructed 3D images was compared between the two methods.
Results: measured artery diameter using a single gantry rotation with constant speed and acquisition rate was 10.4mm. ACROBEAT was able to reduce the measured artery diameter to 8.75mm. Additionally, by adapting the gantry velocity and projection rate of the imaging system with changes in the patient’s ECG, ACROBEAT enabled a 60% reduction in the total imaging dose compared to the standard of care for mid-procedural imaging.
Conclusion: is the first application of our novel adaptive imaging protocol, ACROBEAT, outside of the thoracic region. ACROBEAT has the potential to provide sharper and safer images for intracranial interventional procedures.
Funding Support, Disclosures, and Conflict of Interest: This research was supported by a Siemens grant. The concept and information presented in this paper are based on research and is not commercially available. Due to regulatory reasons its future availability cannot be guaranteed. Also supported by grant #1123068 through the Priority-driven Collaborative Cancer Research Scheme by Cancer Australia.