Visibility of Intracranial Perforating Arteries Using Ultra-High-Resolution Photon-Counting Detector Computed Tomography (CT) Angiography

<b>Background/Objectives:</b> Photon-counting detector computed tomography (PCD-CT) offers energy-resolved CT data with enhanced resolution, reduced electronic noise, and improved tissue contrast. This study aimed to evaluate the visibility of intracranial perforating arteries on ultra-high-resolution (UHR) CT angiography (CTA) on PCD-CT. <b>Methods:</b> A retrospective analysis of intracranial UHR PCD-CTA was performed for 30 patients. The image quality from four UHR PCD-CTA reconstruction methods [kernel Hv40 and Hv72, with and without quantum iterative reconstruction (QIR)] was assessed for the lenticulostriate arteries (LSAs) and pontine arteries (PAs). A subjective evaluation included peripheral visibility, vessel sharpness, and image noise, while objective analysis focused on the signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR). <b>Results:</b> Peripheral LSAs were well visualized across all reconstruction methods, with no significant differences between them. Vessel sharpness and image noise varied significantly (<i>p</i> < 0.0001); sharper LSAs and more noise were seen with kernel Hv72 compared to kernel Hv40 (<i>p</i> < 0.05). A similar pattern was observed for PAs, though peripheral visibility was lower than that for LSAs. The SNR and CNR were the highest in the presence of kernel Hv72 with QIR, and lowest with kernel Hv72 without QIR, compared to kernel Hv40 (<i>p</i> < 0.05). <b>Conclusions:</b> UHR PCD-CTA provided a good visualization of the intracranial perforating arteries, particularly LSAs. The vessel sharpness and image noise varied by reconstruction method, in which kernel Hv72 with QIR offered the optimal visualization.