Related Experiment Video
Updated: May 18, 2026
A Volumetric Method for Quantification of Cerebral Vasospasm in a Murine Model of Subarachnoid Hemorrhage
Published on: July 28, 2018
Comparison of 3D computer-aided with manual cerebral aneurysm measurements in different imaging modalities
M Groth1, N D Forkert, J H Buhk
1Department of Diagnostic and Interventional Neuroradiology, University Medical Center Hamburg-Eppendorf, Martinistr. 52, 20246 Hamburg, Germany. groth.michael@googlemail.com
Introduction:
To compare intra- and inter-observer reliability of aneurysm measurements obtained by a 3D computer-aided technique with standard manual aneurysm measurements in different imaging modalities.
Methods:
A total of 21 patients with 29 cerebral aneurysms were studied. All patients underwent digital subtraction angiography (DSA), contrast-enhanced (CE-MRA) and time-of-flight magnetic resonance angiography (TOF-MRA). Aneurysm neck and depth diameters were manually measured by two observers in each modality. Additionally, semi-automatic computer-aided diameter measurements were performed using 3D vessel surface models derived from CE- (CE-com) and TOF-MRA (TOF-com) datasets. Bland-Altman analysis (BA) and intra-class correlation coefficient (ICC) were used to evaluate intra- and inter-observer agreement.
Results:
BA revealed the narrowest relative limits of intra- and inter-observer agreement for aneurysm neck and depth diameters obtained by TOF-com (ranging between ±5.3 % and ±28.3 %) and CE-com (ranging between ±23.3 % and ±38.1 %). Direct measurements in DSA, TOF-MRA and CE-MRA showed considerably wider limits of agreement. The highest ICCs were observed for TOF-com and CE-com (ICC values, 0.92 or higher for intra- as well as inter-observer reliability).
Conclusion:
Computer-aided aneurysm measurement in 3D offers improved intra- and inter-observer reliability and a reproducible parameter extraction, which may be used in clinical routine and as objective surrogate end-points in clinical trials.
