Evaluation of Vessel Enhancement Software for Improved Cerebrovascular Visualization in 3D Rotational Angiography

3D rotational angiography (3D-RA) is a key technique in endovascular neurointerventions but often provides limited visibility of small, clinically-relevant vessels such as perforating arteries. This study aimed to evaluate the performance of a novel, dedicated vessel-enhancement software in improving vascular visibility in 3D-RA data sets.

In this retrospective analysis of a prospectively maintained institutional database, 50 consecutive patients with unruptured anterior circulation aneurysms underwent standardized 3D-RA acquisitions on the Artis icono angiographic system. Each data set was reconstructed with and without vessel enhancement (VE). Quantitative analysis comprised contrast-to-noise ratio (CNR) measurements in the M1 segment of the MCA and in perforating branches. Qualitative image quality was independently assessed by 2 experienced readers on a 5-point scale.

VE yielded consistent improvements across all analyses. CNR increased in the M1 segment (mean CNR, no VE: 25.15 [SD, 5.42] versus VE: 27.85 [SD 5.41], P < .001) and in the perforating arteries (mean CNR, no vessel enhancement: 22.26 [SD, 4.57] versus VE: 24.80 [SD, 4.75], P < .001). Overall qualitative ratings improved from "very good" to "excellent" diagnostic quality, with higher scores for vessel-enhanced reconstructions compared with unenhanced images (mean score [SD], no VE: 3.90 [SD, 0.65] versus VE: 4.66 [SD, 0.52], P < .001). Improvements were particularly evident for the continuity and visibility of fine vascular structures. Inter-reader agreement was good (κ = 0.72), supporting the robustness of the findings.

Dedicated VE software improves the depiction of both larger arteries and small perforating branches in cerebral 3D-RA. By enabling clearer visualization of fine vascular structures, this approach may strengthen diagnostic confidence and procedural safety in complex endovascular neurointerventions.