Utility of follow-up ultra-high-resolution CT angiography with model-based iterative reconstruction after flow diverter treatment for cerebral aneurysms.

PURPOSE: Follow-up examinations after flow diverter (FD) treatment for cerebral aneurysms typically involve magnetic resonance imaging (MRI) or digital subtraction angiography (DSA). However, MRI is prone to vascular defects due to metal artifacts from FD, and DSA carries a risk of ischemic complications. In the context of computed tomography angiography (CTA), this study compares the efficacy of ultra-high-resolution CT (UHRCT) and novel reconstruction techniques, such as model-based iterative reconstruction (MBIR), against conventional methods such as filtered back projection (FBP) and hybrid iterative reconstruction (IR), to determine if they are a viable alternative to DSA in clinical settings. MATERIALS AND METHODS: A phantom study was conducted with the full-width half-maximum considered as the FD thickness. This study compared three reconstruction methods: MBIR, FBP, and hybrid IR. A clinical study was also conducted with 21 patients who underwent follow-up CTA after FD treatment. The FD's visibility was assessed using a 4-point scale in FBP, hybrid IR, and MBIR compared to cone-beam CT (CBCT) with angiographic systems. RESULTS: In the phantom study, FBP, hybrid IR, and MBIR visualized thinner FD thicknesses and improved detail rendering in that order. MBIR proved to be significantly superior in both the phantom and clinical study. CONCLUSION: UHRCT with MBIR is highly effective for follow-up evaluations after FD treatment and may become the first-choice modality in the future.

[Evaluation of the individual tube current setting in electrocardiogram-gated cardiac computed tomography estimated from plain chest computed tomography using computed tomography automatic exposure control].

The aim of this study was to estimate the tube current on a cardiac computed tomography (CT) from a plain chest CT using CT-automatic exposure control (CT-AEC), to obtain consistent image noise, and to optimize the scan tube current by individualizing the tube current. Sixty-five patients (Group A) underwent cardiac CT at fixed tube current. The mAs value for plain chest CT using CT-AEC (AEC value) and cardiac CT image noise were measured. The tube current needed to obtain the intended level of image noise in the cardiac CT was determined from their correlation. Another 65 patients (Group B) underwent cardiac CT with tube currents individually determined from the AEC value. Image noise was compared among Group A and B. Image noise of cardiac CT in Group B was 24.4 +/- 3.1 HU and was more uniform than in Group A (21.2 +/- 6.1 HU). The error with the desired image noise of 25 HU was lower in Group B (2.4%) than in Group A (15.2%). Individualized tube current selection based on AEC value thus provided consistent image noise and a scan tube current optimized for cardiac CT.