Evaluation of collateral circulation in symptomatic Moyamoya disease using artificial intelligence-assisted whole-brain computed tomography perfusion-derived dynamic four-dimensional computed tomography angiography.
Authors
Affiliations (1)
Affiliations (1)
- Department of Medical Imaging Center, The First Affiliated Hospital of Shihezi University, Shihezi, China.
Abstract
This study aimed to evaluate the clinical value of artificial intelligence-assisted dynamic four-dimensional computed tomography angiography (4D-CTA) reconstructed from whole-brain computed tomography perfusion (CTP) source data for assessing collateral circulation in symptomatic Moyamoya disease (MMD). This retrospective study enrolled 36 patients with digital subtraction angiography (DSA)-confirmed symptomatic MMD who underwent CTP and single-phase CTA (sCTA). Three datasets were derived: group A, conventional sCTA; group B, CTP-derived arterial-phase images; and group C, artificial intelligence-postprocessed 4D-CTA using the Shukun AI algorithm. Image quality was assessed in the arterial phase using objective parameters [CT attenuation, image noise, signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR)] and subjective scores. The diagnostic performance of 4D-CTA was evaluated against DSA for modified Suzuki staging (mSS) and collateral circulation assessment. Group C showed significantly higher CT attenuation value of the vessel, SNR, CNR, and subjective scores, and significantly lower standard deviation of the vessel and standard deviation of the brain parenchyma than groups A and B (all P < 0.05; most P < 0.001). The 4D-CTA showed excellent agreement with DSA in mSS grading (P < 0.001) and a strong positive correlation (P < 0.001). For collateral circulation grading, good agreement (P < 0.001) and moderate positive correlation (P < 0.001) were observed. Artificial intelligence-assisted 4D-CTA reconstructed from whole-brain CTP enables comprehensive multiphase vascular visualization without additional radiation exposure. It facilitates rapid, reliable, noninvasive evaluation of collateral circulation in symptomatic MMD, offering diagnostic accuracy comparable with DSA with promising potential for routine clinical application.