Abstract TP138: Image Quality Assessment of Photon Counting Computed Tomography Angiography of the Head and Neck

• Published in: Stroke (1970) Vol. 55; no. Suppl_1; p. ATP138
• Main Authors: Song, Jae W, Colfer, Michael, Pasyar, Pouyan, Kasner, Scott E, Cucchiara, Brett L, Saba, Luca, Noel, Peter B, Woo, John H
• Format: Journal Article
• Language: English
• Published: Hagerstown, MD Lippincott Williams & Wilkins 01.02.2024
• Subjects: Computed tomography; Angiography; Imaging
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/str.55.suppl_1.TP138

Introduction: Photon-counting computed tomography (PCCT) is a novel advanced CT technology that enables higher spatial resolution, spectral imaging to resolve tissue types and can reduce artifacts. Cerebrovascular image quality from head and neck CTAs acquired on a photon-counting detector was assessed. Methods: Subjects who underwent head and neck PCCTAs for suspected ischemic stroke were retrospectively identified (mean age=57.5 years, N=6 men). Images were acquired with a tube voltage of 120 kVp. Tube current was adjusted based on patient size. Image quality was assessed on a polyenergetic reconstruction. Two neuroradiologists measured mean attenuation [Hounsfield unit (HU)], image noise, and contrast-to-noise ratio (CNR)] of the head/neck arteries. Beam hardening effects on the arteries were calculated using regions of interests (ROIs). CNR was calculated as the absolute difference between the HUmean of the arterial segment and surrounding fat divided by image noise. Intraclass correlation coefficients (ICC) were calculated to measure reliability. Results: Ten head and neck PCCTAs met criteria for diagnostic quality (100% interrater agreement). The CNR of all arterial segments of the extra- and intracranial arteries ranged from 40.7 to 52.7. Effects of beam hardening were worse at the vertebral artery origins, internal carotid artery (ICA)-C3 (cavernous) and ICA-C4 (clinoid) segments. Artifact could be reduced using higher virtual monoenergetic levels, an advantage for segments vulnerable to arterial dissections, such as the vertebral artery origins. Interreader agreement of HUROIs was ICC>0.7 for each arterial segment. Conclusions: Vertebral artery origins near the arch in the neck and cavernous/clinoid ICA segments in the head showed the greatest need for optimization to reduce artifacts and image noise. Future work will focus on leveraging PCCT technology to select the optimal energy level, reduce artifacts and improve diagnostic accuracy.