Can we explain machine learning-based prediction for rupture status assessments of intracranial aneurysms?

Although applying machine learning (ML) algorithms to rupture status assessment of intracranial aneurysms (IA) has yielded promising results, the opaqueness of some ML methods has limited their clinical translation. We presented the first explainability comparison of six commonly used ML algorithms: multivariate logistic regression (LR), support vector machine (SVM), random forest (RF), extreme gradient boosting (XGBoost), multi-layer perceptron neural network (MLPNN), and Bayesian additive regression trees (BART). A total of 112 IAs with known rupture status were selected for this study. The ML-based classification used two anatomical features, nine hemodynamic parameters, and thirteen morphologic variables. We utilized permutation feature importance, local interpretable model-agnostic explanations (LIME), and SHapley Additive exPlanations (SHAP) algorithms to explain and analyze 6 Ml algorithms. All models performed comparably: LR area under the curve (AUC) was 0.71; SVM AUC was 0.76; RF AUC was 0.73; XGBoost AUC was 0.78; MLPNN AUC was 0.73; BART AUC was 0.73. Our interpretability analysis demonstrated consistent results across all the methods; i.e., the utility of the top 12 features was broadly consistent. Furthermore, contributions of 9 important features (aneurysm area, aneurysm location, aneurysm type, wall shear stress maximum during systole, ostium area, the size ratio between aneurysm width, (parent) vessel diameter, one standard deviation among time-averaged low shear area, and one standard deviation of temporally averaged low shear area less than 0.4 Pa) were nearly the same. This research suggested that ML classifiers can provide explainable predictions consistent with general domain knowledge concerning IA rupture. With the improved understanding of ML algorithms, clinicians' trust in ML algorithms will be enhanced, accelerating their clinical translation.

Artificial Intelligence-Based 3D Angiography for Visualization of Complex Cerebrovascular Pathologies.

BACKGROUND AND PURPOSE: By means of artificial intelligence, 3D angiography is a novel postprocessing method for 3D imaging of cerebral vessels. Because 3D angiography does not require a mask run like the current standard 3D-DSA, it potentially offers a considerable reduction of the patient radiation dose. Our aim was an assessment of the diagnostic value of 3D angiography for visualization of cerebrovascular pathologies. MATERIALS AND METHODS: 3D-DSA data sets of cerebral aneurysms (n CA = 10), AVMs (n AVM = 10), and dural arteriovenous fistulas (dAVFs) (n dAVF = 10) were reconstructed using both conventional and prototype software. Corresponding reconstructions have been analyzed by 2 neuroradiologists in a consensus reading in terms of image quality, injection vessel diameters (vessel diameter [VD] 1/2), vessel geometry index (VGI = VD1/VD2), and specific qualitative/quantitative parameters of AVMs (eg, location, nidus size, feeder, associated aneurysms, drainage, Spetzler-Martin score), dAVFs (eg, fistulous point, main feeder, diameter of the main feeder, drainage), and cerebral aneurysms (location, neck, size). RESULTS: In total, 60 volumes have been successfully reconstructed with equivalent image quality. The specific qualitative/quantitative assessment of 3D angiography revealed nearly complete accordance with 3D-DSA in AVMs (eg, mean nidus size3D angiography/3D-DSA= 19.9 [SD, 10.9]/20.2 [SD, 11.2] mm; r = 0.9, P = .001), dAVFs (eg, mean diameter of the main feeder3D angiography/3D-DSA= 2.04 [SD, 0.65]/2.05 [SD, 0.63] mm; r = 0.9, P = .001), and cerebral aneurysms (eg, mean size3D angiography/3D-DSA= 5.17 [SD, 3.4]/5.12 [SD, 3.3] mm; r = 0.9, P = .001). Assessment of the geometry of the injection vessel in 3D angiography data sets did not differ significantly from that of 3D-DSA (vessel geometry indexAVM: r = 0.84, P = .003; vessel geometry indexdAVF: r = 0.82, P = .003; vessel geometry indexCA: r = 0.84, P <.001). CONCLUSIONS: In this study, the artificial intelligence-based 3D angiography was a reliable method for visualization of complex cerebrovascular pathologies and showed results comparable with those of 3D-DSA. Thus, 3D angiography is a promising postprocessing method that provides a significant reduction of the patient radiation dose.

4D-DSA: Development and Current Neurovascular Applications.

Originally described by Davis et al in 2013, 4D-Digital Subtraction Angiography (4D-DSA) has developed into a commercially available application of DSA in the angiography suite. 4D-DSA provides the user with 3D time-resolved images, allowing observation of a contrast bolus at any desired viewing angle through the vasculature and at any time point during the acquisition (any view at any time). 4D-DSA mitigates some limitations that are intrinsic to both 2D- and 3D-DSA images. The clinical applications for 4D-DSA include evaluations of AVMs and AVFs, intracranial aneurysms, and atherosclerotic occlusive disease. Recent advances in blood flow quantification using 4D-DSA indicate that these data provide both the velocity and geometric information necessary for the quantification of blood flow. In this review, we will discuss the development, acquisition, reconstruction, and current neurovascular applications of 4D-DSA volumes.

Optimizing the Quality of 4D-DSA Temporal Information.

BACKGROUND AND PURPOSE: Quantification of blood flow using a 4D-DSA would be useful in the diagnosis and treatment of cerebrovascular diseases. A protocol optimizing identification of density variations in the time-density curves of a 4D-DSA has not been defined. Our purpose was to determine the contrast injection protocol most likely to result in the optimal pulsatility signal strength. MATERIALS AND METHODS: Two 3D-printed patient-specific models were used and connected to a pulsatile pump and flow system, which delivered 250-260 mL/min to the model. Contrast medium (Isovue, 370 mg I/mL, 75% dilution) was injected through a 6F catheter positioned upstream from the inlet of the model. 4D-DSA acquisitions were performed for the following injection rates: 1.5, 2.0, 2.5, 3.0 and 3.5 mL/s for 8 seconds. To determine pulsatility, we analyzed the time-density curve at the inlets using the oscillation amplitude and a previously described numeric metric, the sideband ratio. Vascular geometry from 4D-DSA reconstructions was compared with ground truth and micro-CT measurements of the model. Dimensionless numbers that characterize hemodynamics, Reynolds and Craya-Curtet, were calculated for each injection rate. RESULTS: The strongest pulsatility signal occurred with the 2.5 mL/s injections. The largest oscillation amplitudes were found with 2.0- and 2.5-mL/s injections. Geometric accuracy was best preserved with injection rates of >1.5 mL/s. CONCLUSIONS: An injection rate of 2.5 mL/s provided the strongest pulsatility signal in the 4D-DSA time-density curve. Geometric accuracy was best preserved with injection rates above 1.5 mL/s. These results may be useful in future in vivo studies of blood flow quantification.

Quantitative and Qualitative Comparison of 4D-DSA with 3D-DSA Using Computational Fluid Dynamics Simulations in Cerebral Aneurysms.

BACKGROUND AND PURPOSE: 4D-DSA allows time-resolved 3D imaging of the cerebral vasculature. The aim of our study was to evaluate this method in comparison with the current criterion standard 3D-DSA by qualitative and quantitative means using computational fluid dynamics. MATERIALS AND METHODS: 3D- and 4D-DSA datasets were acquired in patients with cerebral aneurysms. Computational fluid dynamics analysis was performed for all datasets. Using computational fluid dynamics, we compared 4D-DSA with 3D-DSA in terms of both aneurysmal geometry (quantitative: maximum diameter, ostium size [OZ1/2], volume) and hemodynamic parameters (qualitative: flow stability, flow complexity, inflow concentration; quantitative: average/maximum wall shear stress, impingement zone, low-stress zone, intra-aneurysmal pressure, and flow velocity). Qualitative parameters were descriptively analyzed. Correlation coefficients (r, P value) were calculated for quantitative parameters. RESULTS: 3D- and 4D-DSA datasets of 10 cerebral aneurysms in 10 patients were postprocessed. Evaluation of aneurysmal geometry with 4D-DSA (r maximum diameter = 0.98, P maximum diameter <.001; r OZ1/OZ2 = 0.98/0.86, P OZ1/OZ2 < .001/.002; r volume = 0.98, P volume <.001) correlated highly with 3D-DSA. Evaluation of qualitative hemodynamic parameters (flow stability, flow complexity, inflow concentration) did show complete accordance, and evaluation of quantitative hemodynamic parameters (r average/maximum wall shear stress diastole = 0.92/0.88, P average/maximum wall shear stress diastole < .001/.001; r average/maximum wall shear stress systole = 0.94/0.93, P average/maximum wall shear stress systole < .001/.001; r impingement zone = 0.96, P impingement zone < .001; r low-stress zone = 1.00, P low-stress zone = .01; r pressure diastole = 0.84, P pressure diastole = .002; r pressure systole = 0.9, P pressure systole < .001; r flow velocity diastole = 0.95, P flow velocity diastole < .001; r flow velocity systole = 0.93, P flow velocity systole < .001) did show nearly complete accordance between 4D- and 3D-DSA. CONCLUSIONS: Despite a different injection protocol, 4D-DSA is a reliable basis for computational fluid dynamics analysis of the intracranial vasculature and provides equivalent visualization of aneurysm geometry compared with 3D-DSA.

Quantification of Blood Velocity with 4D Digital Subtraction Angiography Using the Shifted Least-Squares Method.

BACKGROUND AND PURPOSE: 4D-DSA provides time-resolved 3D-DSA volumes with high temporal and spatial resolutions. The purpose of this study is to investigate a shifted least squares method to estimate the blood velocity from the 4D DSA images. Quantitative validation was performed using a flow phantom with an ultrasonic flow probe as ground truth. Quantification of blood velocity in human internal carotid arteries was compared with measurements generated from 3D phase-contrast MR imaging. MATERIALS AND METHODS: The centerlines of selected vascular segments and the time concentration curves of each voxel along the centerlines were determined from the 4D-DSA dataset. The temporal shift required to achieve a minimum difference between any point and other points along the centerline of a segment was calculated. The temporal shift as a function of centerline point position was fit to a straight line to generate the velocity. The proposed shifted least-squares method was first validated using a flow phantom study. Blood velocities were also estimated in the 14 ICAs of human subjects who had both 4D-DSA and phase-contrast MR imaging studies. Linear regression and correlation analysis were performed on both the phantom study and clinical study, respectively. RESULTS: Mean velocities of the flow phantom calculated from 4D-DSA matched very well with ultrasonic flow probe measurements with 11% relative root mean square error. Mean blood velocities of ICAs calculated from 4D-DSA correlated well with phase-contrast MR imaging measurements with Pearson correlation coefficient r = 0.835. CONCLUSIONS: The availability of 4D-DSA provides the opportunity to use the shifted least-squares method to estimate velocity in vessels within a 3D volume.

3D Deep Learning Angiography (3D-DLA) from C-arm Conebeam CT.

BACKGROUND AND PURPOSE: Deep learning is a branch of artificial intelligence that has demonstrated unprecedented performance in many medical imaging applications. Our purpose was to develop a deep learning angiography method to generate 3D cerebral angiograms from a single contrast-enhanced C-arm conebeam CT acquisition in order to reduce image artifacts and radiation dose. MATERIALS AND METHODS: A set of 105 3D rotational angiography examinations were randomly selected from an internal data base. All were acquired using a clinical system in conjunction with a standard injection protocol. More than 150 million labeled voxels from 35 subjects were used for training. A deep convolutional neural network was trained to classify each image voxel into 3 tissue types (vasculature, bone, and soft tissue). The trained deep learning angiography model was then applied for tissue classification into a validation cohort of 8 subjects and a final testing cohort of the remaining 62 subjects. The final vasculature tissue class was used to generate the 3D deep learning angiography images. To quantify the generalization error of the trained model, we calculated the accuracy, sensitivity, precision, and Dice similarity coefficients for vasculature classification in relevant anatomy. The 3D deep learning angiography and clinical 3D rotational angiography images were subjected to a qualitative assessment for the presence of intersweep motion artifacts. RESULTS: Vasculature classification accuracy and 95% CI in the testing dataset were 98.7% (98.3%-99.1%). No residual signal from osseous structures was observed for any 3D deep learning angiography testing cases except for small regions in the otic capsule and nasal cavity compared with 37% (23/62) of the 3D rotational angiographies. CONCLUSIONS: Deep learning angiography accurately recreated the vascular anatomy of the 3D rotational angiography reconstructions without a mask. Deep learning angiography reduced misregistration artifacts induced by intersweep motion, and it reduced radiation exposure required to obtain clinically useful 3D rotational angiography.

4D DSA for Dynamic Visualization of Cerebral Vasculature: A Single-Center Experience in 26 Cases.

BACKGROUND AND PURPOSE: 4D DSA allows acquisition of time-resolved 3D reconstructions of cerebral vessels by using C-arm conebeam CT systems. The aim of our study was to evaluate this new method by qualitative and quantitative means. MATERIALS AND METHODS: 2D and 4D DSA datasets were acquired in patients presenting with AVMs, dural arteriovenous fistulas, and cerebral aneurysms. 4D DSA was compared with 2D DSA in a consensus reading of qualitative and quantitative parameters of AVMs (eg, location, feeder, associated aneurysms, nidus size, drainage, Martin-Spetzler Score), dural arteriovenous fistulas (eg, fistulous point, main feeder, diameter of the main feeder, drainage), and cerebral aneurysms (location, neck configuration, aneurysmal size). Identifiability of perforators and diameters of the injection vessel (ICA, vertebral artery) were analyzed in 2D and 4D DSA. Correlation coefficients and a paired t test were calculated for quantitative parameters. The effective patient dose of the 4D DSA protocol was evaluated with an anthropomorphic phantom. RESULTS: In 26 patients, datasets were acquired successfully (AVM = 10, cerebral aneurysm = 10, dural arteriovenous fistula = 6). Qualitative and quantitative evaluations of 4D DSA in AVMs (nidus size: r = 0.99, P = .001), dural arteriovenous fistulas (diameter of the main feeder: r = 0.954, P = .03), and cerebral aneurysms (aneurysmal size: r = 1, P = .001) revealed nearly complete accordance with 2D DSA. Perforators were comparably visualized with 4D DSA. Measurement of the diameter of the injection vessel in 4D DSA was equivalent to that in 2D DSA (P = .039). The effective patient dose of 4D DSA was 1.2 mSv. CONCLUSIONS: 4D DSA is feasible for imaging of AVMs, dural arteriovenous fistulas, and cerebral aneurysms. 4D DSA offers reliable visualization of the cerebral vasculature and may improve the understanding and treatment of AVMs and dural arteriovenous fistulas. The number of 2D DSA acquisitions required for an examination may be reduced through 4D DSA.

Comparison of the Diagnostic Utility of 4D-DSA with Conventional 2D- and 3D-DSA in the Diagnosis of Cerebrovascular Abnormalities.

BACKGROUND AND PURPOSE: 4D-DSA is a time-resolved technique that allows viewing of a contrast bolus at any time and from any desired viewing angle. Our hypothesis was that the information content in a 4D-DSA reconstruction was essentially equivalent to that in a combination of 2D acquisitions and a 3D-DSA reconstruction. MATERIALS AND METHODS: Twenty-six consecutive patients who had both 2D- and 3D-DSA acquisitions were included in the study. The angiography report was used to obtain diagnoses and characteristics of abnormalities. Diagnoses included AVM/AVFs, aneurysms, stenosis, and healthy individuals. 4D-DSA reconstructions were independently reviewed by 3 experienced observers who had no part in the clinical care. Using an electronic evaluation form, these observers recorded their assessments based only on the 4D reconstructions. The clinical evaluations were then compared with the 4D evaluations for diagnosis and lesion characteristics. RESULTS: Results showed both interrater and interclass agreements (κ = 0.813 and 0.858). Comparing the 4D diagnosis with the clinical diagnosis for the 3 observers yielded κ values of 0.906, 0.912, and 0.906. The κ values for agreement among the 3 observers for the type of abnormality were 0.949, 0.845, and 0.895. There was complete agreement on the presence of an abnormality between the clinical and 4D-DSA in 23/26 cases. In 2 cases, there were conflicting opinions. CONCLUSIONS: In this study, the information content of 4D-DSA reconstructions was largely equivalent to that of the combined 2D/3D studies. The availability of 4D-DSA should reduce the requirement for 2D-DSA acquisitions.

Evaluation of Collaterals and Clot Burden Using Time-Resolved C-Arm Conebeam CT Angiography in the Angiography Suite: A Feasibility Study.

BACKGROUND AND PURPOSE: The assessment of collaterals and clot burden in patients with acute ischemic stroke provides important information about treatment options and clinical outcome. Time-resolved C-arm conebeam CT angiography has the potential to provide accurate and reliable evaluations of collaterals and clot burden in the angiographic suite. Experience with this technique is extremely limited, and feasibility studies are needed to validate this technique. Our purpose was to present such a feasibility study. MATERIALS AND METHODS: Ten C-arm conebeam CT perfusion datasets from 10 subjects with acute ischemic stroke acquired before endovascular treatment were retrospectively processed to generate time-resolved conebeam CTA. From time-resolved conebeam CTA, 2 experienced readers evaluated the clot burden and collateral flow in consensus by using previously reported scoring systems and assessed the clinical value of this novel imaging technique independently. Interobserver agreement was analyzed by using the intraclass correlation analysis method. RESULTS: Clot burden and collateral flow can be assessed by using the commonly accepted scoring systems for all eligible cases. Additional clinical information (eg, the quantitative dynamic information of collateral flow) can be obtained from this new imaging technique. Two readers agreed that time-revolved C-arm conebeam CTA is the preferred method for evaluating the clot burden and collateral flow compared with other conventional imaging methods. CONCLUSIONS: Comprehensive evaluations of clot burden and collateral flow are feasible by using time-resolved C-arm conebeam CTA data acquired in the angiography suite. This technique further enriches the imaging tools in the angiography suite to enable a "one-stop- shop" imaging workflow for patients with acute ischemic stroke.

C-Arm Conebeam CT Perfusion Imaging in the Angiographic Suite: A Comparison with Multidetector CT Perfusion Imaging.

BACKGROUND AND PURPOSE: Perfusion imaging in the angiography suite may provide a way to reduce time from stroke onset to endovascular revascularization of patients with large-vessel occlusion. Our purpose was to compare conebeam CT perfusion with multidetector CT perfusion. MATERIALS AND METHODS: Data from 7 subjects with both multidetector CT perfusion and conebeam CT perfusion were retrospectively processed and analyzed. Two algorithms were used to enhance temporal resolution and temporal sampling density and reduce the noise of conebeam CT data before generating perfusion maps. Two readers performed qualitative image-quality evaluation on maps by using a 5-point scale. ROIs indicating CBF/CBV abnormalities were drawn. Quantitative analyses were performed by using the Sørensen-Dice coefficients to quantify the similarity of abnormalities. A noninferiority hypothesis was tested to compare conebeam CT perfusion against multidetector CT perfusion. RESULTS: Average image-quality scores for multidetector CT perfusion and conebeam CT perfusion images were 2.4 and 2.3, respectively. The average confidence score in diagnosis was 1.4 for both multidetector CT and conebeam CT; the average confidence scores for the presence of a CBV/CBF mismatch were 1.7 (κ = 0.50) and 1.5 (κ = 0.64). For multidetector CT perfusion and conebeam CT perfusion maps, the average scores of confidence in making treatment decisions were 1.4 (κ = 0.79) and 1.3 (κ = 0.90). The area under the visual grading characteristic for the above 4 qualitative quality scores showed an average area under visual grading characteristic of 0.50, with 95% confidence level cover centered at the mean for both readers. The Sørensen-Dice coefficient for CBF maps was 0.81, and for CBV maps, 0.55. CONCLUSIONS: After postprocessing methods were applied to enhance image quality for conebeam CT perfusion maps, the conebeam CT perfusion maps were not inferior to those generated from multidetector CT perfusion.

Low-Dose Volume-of-Interest C-Arm CT Imaging of Intracranial Stents and Flow Diverters.

BACKGROUND AND PURPOSE: Volume-of-interest C-arm CT is a novel technique for imaging of intracranial high-contrast objects. We performed this study to evaluate the potential diagnostic value and radiation dose reduction of this technique for imaging of intracranial stents and flow diverters. MATERIALS AND METHODS: Twenty-seven patients were imaged with a VOI C-arm CT scan following treatment with a flow diverter or stent-assisted coiling. The radiation dose-area product was recorded for VOI scans. For comparison, the dose-area product from 30 previously acquired consecutive full-view DynaCTs was used. Thermoluminescence dosimetry by using 35 evenly distributed thermoluminescence dosimeters in an anthropomorphic head phantom was also performed by using both conventional full field and VOI acquisitions. Three observers were presented with VOI images for assessment of the potential diagnostic value. RESULTS: The dose-area product measurements showed an exposure reduction of 85% compared with the full field acquisitions used for comparison. The thermoluminescence dosimetry evaluations also showed a considerable dose reduction of 79.8% throughout the volume. For most of the evaluated cases, the observers thought that diagnostically useful information was provided by the VOI images (α = .810). Visualization of device details, such as the extent of opening, positioning, wall apposition, and aneurysm coverage, was judged of good diagnostic quality for most cases (88.9%-92.6%). CONCLUSIONS: In this study, VOI C-arm CT provided high-quality diagnostic images of intracranial stents and flow diverters at a dramatic reduction of radiation exposure. Image content was thought to add useful information. It is a promising method to assess device status during procedures and at follow-up.

Dynamic Angiography and Perfusion Imaging Using Flat Detector CT in the Angiography Suite: A Pilot Study in Patients with Acute Middle Cerebral Artery Occlusions.

BACKGROUND AND PURPOSE: Perfusion and angiographic imaging using intravenous contrast application to evaluate stroke patients is now technically feasible by flat detector CT performed by the angiographic system. The aim of this pilot study was to show the feasibility and qualitative comparability of a novel flat detector CT dynamic perfusion and angiographic imaging protocol in comparison with a multimodal stroke MR imaging protocol. MATERIALS AND METHODS: In 12 patients with acute stroke, MR imaging and the novel flat detector CT protocol were performed before endovascular treatment. Perfusion parameter maps (MTT, TTP, CBV, CBF) and MIP/volume-rendering technique images obtained by using both modalities (MR imaging and flat detector CT) were compared. RESULTS: Comparison of MIP/volume-rendering technique images demonstrated equivalent visibility of the occlusion site. Qualitative comparison of perfusion parameter maps by using ASPECTS revealed high Pearson correlation coefficients for parameters CBF, MTT, and TTP (0.95-0.98), while for CBV, the coefficient was lower (0.49). CONCLUSIONS: We have shown the feasibility of a novel dynamic flat detector CT perfusion and angiographic protocol for the diagnosis and triage of patients with acute ischemic stroke. In a qualitative comparison, the parameter maps and MIP/volume-rendering technique images compared well with MR imaging. In our opinion, this flat detector CT application may substitute for multisection CT imaging in selected patients with acute stroke so that in the future, patients with acute stroke may be directly referred to the angiography suite, thereby avoiding transportation and saving time.

A Comparison of 4D DSA with 2D and 3D DSA in the Analysis of Normal Vascular Structures in a Canine Model.

BACKGROUND AND PURPOSE: 4D DSA allows viewing of 3D DSA as a series of time-resolved volumes of a contrast bolus. There is no comparison of the accuracy of the anatomic information provided by 4D DSA with that available from conventional 2D and 3D DSA. Our purpose was to make this comparison by using a canine model. MATERIALS AND METHODS: 2D, 3D, and 4D DSA acquisitions were performed in 5 canines from 3 catheter positions in the common carotid artery yielding 15 2D, 15 3D, and 15 4D datasets. For each territory, 3 vascular segments were chosen for comparison. Images were reviewed by 2 experienced neuroradiologists and were graded by the ability to visualize a segment, its filling direction, and preferred technique. Two visualization modes for 4D DSA were compared (volume-rendering technique and MIP). RESULTS: 4D DSA was preferred in 73.9% of the image sets; 2D, in 22.7%; and 3D, in 3.4%. 4D DSA MIP rendering yielded superior visualization of very small vessel details; the 4D DSA volume-rendering technique offered superior depth and overlap information and better visualization of the surface details of the vasculature. CONCLUSIONS: In this study, 4D DSA was preferred over 2D and 3D DSA for analysis of normal vasculature. The ability to provide any view of a vascular territory at any time during passage of a contrast bolus seems likely to reduce the need for many 2D acquisitions during diagnostic and therapeutic procedures. This then potentially translates into a reduction in radiation and contrast dose.

Angiographic C-arm CT visualization of the Woven EndoBridge cerebral aneurysm embolization device (WEB): first experience in an animal aneurysm model.

INTRODUCTION: Improvements in the imaging capabilities of angiographic C-arm computed tomography (CT) using flat detector angiographic (FD-CT) systems now provide a means for the in vivo visualization of devices used for the treatment of intracranial aneurysms. One such device, the WEB embolization device, is made of a braided mesh of nitinol wires, the size of which are near to the limits of visualization using conventional x-ray fluoroscopy. Our hypothesis was that the imaging of these implants C-arm CT would provide useful information regarding their positioning and deployment. METHODS: In five New Zealand white rabbits elastase induced aneurysms were created and subsequently treated using a WEB. Imaging was performed using digital subtraction angiography (DSA), X-ray imaging and two different Angiographic C-arm CT protocols. The images were evaluated by two neuroradiologists using an evaluation scale. RESULTS: The mesh of the WEB was barely visible on the DSA or x-ray fluoroscopy images. Volume rendering technique (VRT) reconstruction and multiplanar reconstruction (MPR) of images done using the C-arm CT protocols clearly delineated the shape and structure of the device. Contrast-enhanced MPR and VRT reconstructions allowed assessment of the status of blood flow in the aneurysms. Beam hardening artifacts caused by platinum markers on the WEB were present. CONCLUSION: In vivo C-arm CT imaging of the WEB is feasible and allows precise determination of the position and deployment status of the device. On contrast-enhanced images the occlusion status of aneurysms and the positioning of the WEB in relationship to the parent artery can be evaluated. C-arm-CT may serve as a minimal-invasive follow-up imaging modality.

C-arm CT measurement of cerebral blood volume and cerebral blood flow using a novel high-speed acquisition and a single intravenous contrast injection.

BACKGROUND AND PURPOSE: Assessment of perfusion parameters is important in the selection of patients who are most likely to benefit from revascularization after an acute ischemic stroke. The aim of this study was to evaluate the feasibility of measuring cerebral perfusion parameters with the use of a novel high-speed C-arm CT acquisition in conjunction with a single intravenous injection of contrast. MATERIALS AND METHODS: Seven canines had experimentally induced focal ischemic regions confirmed by CT perfusion imaging. Four hours after ischemic injury creation, each subject underwent cerebral perfusion measurements with the use of standard perfusion CT, immediately followed by the use of C-arm CT. Cerebral blood flow and cerebral blood volume maps measured by C-arm CT were quantitatively and qualitatively compared with those measured by perfusion CT for 6 of the 7 canine subjects. RESULTS: Results from independent observer evaluations of perfusion CT and C-arm perfusion maps show strong agreement between observers for identification of ischemic lesion location. Significant percentage agreement between observers for lesion detection and identification of perfusion mismatch between CBV and CBF maps indicate that the maps for both perfusion CT and C-arm are easy to interpret. Quantitative region of interest-based evaluation showed a strong correlation between the perfusion CT and C-arm CBV and CBF maps (R(2) = 0.68 and 0.85). C-arm measurements for both CBV and CBF were consistently overestimated when compared with perfusion CT. CONCLUSIONS: Qualitative and quantitative measurements of CBF and CBV with the use of a C-arm CT acquisition and a single intravenous injection of contrast agent are feasible. Future improvements in flat detector technology and software algorithms probably will enable more accurate quantitative perfusion measurements with the use of C-arm CT.

4D digital subtraction angiography: implementation and demonstration of feasibility.

BACKGROUND AND PURPOSE: Conventional 3D-DSA volumes are reconstructed from a series of projections containing temporal information. It was our purpose to develop a technique which would generate fully time-resolved 3D-DSA vascular volumes having better spatial and temporal resolution than that which is available with CT or MR angiography. MATERIALS AND METHODS: After a single contrast injection, projections from the mask and fill rotation are subtracted to create a series of vascular projections. With the use of these projections, a conventional conebeam CT reconstruction is generated (conventional 3D-DSA). This is used to constrain the reconstruction of individual 3D temporal volumes, which incorporate temporal information from the acquired projections (4D-DSA). RESULTS: Typically, 30 temporal volumes per second are generated with the use of currently available flat detector systems, a factor of ∼200 increase over that achievable with the use of multiple gantry rotations. Dynamic displays of the reconstructed volumes are viewable from any angle. Good results have been obtained by using both intra-arterial and intravenous injections. CONCLUSIONS: It is feasible to generate time-resolved 3D-DSA vascular volumes with the use of commercially available flat detector angiographic systems and clinically practical injection protocols. The spatial resolution and signal-to-noise ratio of the time frames are largely determined by that of the conventional 3D-DSA constraining image and not by that of the projections used to generate the 3D reconstruction. The spatial resolution and temporal resolution exceed that of CTA and MRA, and the small vessel contrast is increased relative to that of conventional 2D-DSA due to the use of maximum intensity projections.

Monitoring of balloon test occlusion of the internal carotid artery by parametric color coding and perfusion imaging within the angio suite: first results.

BACKGROUND: Temporary balloon test occlusion (BTO) might be performed prior to procedures in which occlusion of the internal carotid artery (ICA) might be necessary. We tested the hypothesis that parametric color coding (PCC) of angiographic series (digital subtraction angiography (DSA)) along with the assessment of cerebral blood volume (CBV) in the angiography suite would simplify and enhance the identification of candidates who are most likely to tolerate occlusion. MATERIALS AND METHODS: Fifteen patients underwent angiographic series (DSA) and perfusion imaging before and during BTO. Pre- and postocclusion DSA acquisitions were evaluated for venous delay by conventional methods ("eye balling") and by PCC measurements. Comparison of CBV values between the left and right hemisphere in 6 defined regions was performed. RESULTS: Values of venous delay by eye balling and PCC showed a high correlation (r = 0.87, p < 0.01). Bland-Altman plot indicated slightly lower values (-0.05 s) by the PCC method. One of the 15 patients developed an asymmetrical CBV map with an increase in CBV of more than one standard deviation in 3 of the 6 regions of interest (ROIs). Acquisition of angiographic series and perfusion imaging did not prolong the test occlusion time. CONCLUSION: PCC and CBV mapping are feasible during BTO. The use of PCC seems to simplify the ability to measure changes in venous filling delay. Perfusion imaging may show an increase in CBV in patients reaching the limits of cerebral autoregulation. These patients may be at risk for delayed infarction, even though they seem to tolerate temporary occlusion, and could be unsuitable candidates for permanent ICA occlusion.

WEB device for endovascular treatment of wide-neck bifurcation aneurysms.

BACKGROUND AND PURPOSE: The WEB is an intrasaccular flow disrupter dedicated to EVT of IA. We report our initial experience in a series of patients treated with this device. MATERIALS AND METHODS: This prospective study was approved by the authors' ethical committees. Nineteen patients with 20 unruptured wide-neck bifurcation IAs were treated by WEB placement. Technical issues, immediate posttreatment angiographic findings, and clinical and imaging follow-up at 3, 6, and 12 months were assessed. RESULTS: Failure of WEB placement occurred in 1 case because of unavailability of a suitably sized device. Embolization was successful in 18 patients with 19 IAs, and it required additional stent placement and/or coiling in 3 cases at the acute phase and in 1 case at follow-up. Two patients experienced a symptomatic complication, and 16 patients had normal neurologic examination findings at discharge. Immediate anatomic outcome showed 1 complete occlusion, 13 near-complete occlusions, and 5 incomplete occlusions. At follow-up, 17 patients had normal neurologic examination findings and 1 retained a hemiparesis. Angiographic controls were obtained in all patients (mean, 6 months), and they showed stable or improved results in all except 4 cases, including 2 complete occlusions, 15 near-complete occlusions, and 2 incomplete occlusions. CONCLUSIONS: In this initial series of patients, EVT of wide-neck bifurcation IAa with the WEB was feasible. Further studies are needed to evaluate the indications, safety, and efficacy of this new technique.