Does the DSA reconstruction kernel affect hemodynamic predictions in intracranial aneurysms? An analysis of geometry and blood flow variations.

BACKGROUND: Computational fluid dynamics (CFD) blood flow predictions in intracranial aneurysms promise great potential to reveal patient-specific flow structures. Since the workflow from image acquisition to the final result includes various processing steps, quantifications of the individual introduced potential error sources are required. METHODS: Three-dimensional (3D) reconstruction of the acquired imaging data as input to 3D model generation was evaluated. Six different reconstruction modes for 3D digital subtraction angiography (DSA) acquisitions were applied to eight patient-specific aneurysms. Segmentations were extracted to compare the 3D luminal surfaces. Time-dependent CFD simulations were carried out in all 48 configurations to assess the velocity and wall shear stress (WSS) variability due to the choice of reconstruction kernel. RESULTS: All kernels yielded good segmentation agreement in the parent artery; deviations of the luminal surface were present at the aneurysm neck (up to 34.18%) and in distal or perforating arteries. Observations included pseudostenoses as well as noisy surfaces, depending on the selected reconstruction kernel. Consequently, the hemodynamic predictions show a mean SD of 11.09% for the aneurysm neck inflow rate, 5.07% for the centerline-based velocity magnitude, and 17.83%/9.53% for the mean/max aneurysmal WSS, respectively. In particular, vessel sections distal to the aneurysms yielded stronger variations of the CFD values. CONCLUSIONS: The choice of reconstruction kernel for DSA data influences the segmentation result, especially for small arteries. Therefore, if precise morphology measurements or blood flow descriptions are desired, a specific reconstruction setting is required. Furthermore, research groups should be encouraged to denominate the kernel types used in future hemodynamic studies.

A 2D driven 3D vessel segmentation algorithm for 3D digital subtraction angiography data.

Cerebrovascular disease is among the leading causes of death in western industrial nations. 3D rotational angiography delivers indispensable information on vessel morphology and pathology. Physicians make use of this to analyze vessel geometry in detail, i.e. vessel diameters, location and size of aneurysms, to come up with a clinical decision. 3D segmentation is a crucial step in this pipeline. Although a lot of different methods are available nowadays, all of them lack a method to validate the results for the individual patient. Therefore, we propose a novel 2D digital subtraction angiography (DSA)-driven 3D vessel segmentation and validation framework. 2D DSA projections are clinically considered as gold standard when it comes to measurements of vessel diameter or the neck size of aneurysms. An ellipsoid vessel model is applied to deliver the initial 3D segmentation. To assess the accuracy of the 3D vessel segmentation, its forward projections are iteratively overlaid with the corresponding 2D DSA projections. Local vessel discrepancies are modeled by a global 2D/3D optimization function to adjust the 3D vessel segmentation toward the 2D vessel contours. Our framework has been evaluated on phantom data as well as on ten patient datasets. Three 2D DSA projections from varying viewing angles have been used for each dataset. The novel 2D driven 3D vessel segmentation approach shows superior results against state-of-the-art segmentations like region growing, i.e. an improvement of 7.2% points in precision and 5.8% points for the Dice coefficient. This method opens up future clinical applications requiring the greatest vessel accuracy, e.g. computational fluid dynamic modeling.

Classification-based summation of cerebral digital subtraction angiography series for image post-processing algorithms.

X-ray-based 2D digital subtraction angiography (DSA) plays a major role in the diagnosis, treatment planning and assessment of cerebrovascular disease, i.e. aneurysms, arteriovenous malformations and intracranial stenosis. DSA information is increasingly used for secondary image post-processing such as vessel segmentation, registration and comparison to hemodynamic calculation using computational fluid dynamics. Depending on the amount of injected contrast agent and the duration of injection, these DSA series may not exhibit one single DSA image showing the entire vessel tree. The interesting information for these algorithms, however, is usually depicted within a few images. If these images would be combined into one image the complexity of segmentation or registration methods using DSA series would drastically decrease. In this paper, we propose a novel method automatically splitting a DSA series into three parts, i.e. mask, arterial and parenchymal phase, to provide one final image showing all important vessels with less noise and moving artifacts. This final image covers all arterial phase images, either by image summation or by taking the minimum intensities. The phase classification is done by a two-step approach. The mask/arterial phase border is determined by a Perceptron-based method trained from a set of DSA series. The arterial/parenchymal phase border is specified by a threshold-based method. The evaluation of the proposed method is two-sided: (1) comparison between automatic and medical expert-based phase selection and (2) the quality of the final image is measured by gradient magnitudes inside the vessels and signal-to-noise (SNR) outside. Experimental results show a match between expert and automatic phase separation of 93%/50% and an average SNR increase of up to 182% compared to summing up the entire series.

New X-ray imaging modalities and their integration with intravascular imaging and interventions.

During recent years various techniques emerged providing more detailed images and insights in the cardiovascular system. C-Arm computed tomography is currently introduced in cardiac imaging offering the potential of three dimensional imaging of the coronary arteries, the cardiac chambers, venous system and a variety of anatomic anomalies inside the interventional environment. Furthermore it might enable perfusion imaging during percutaneous coronary intervention (PCI). Intravascular ultrasound (IVUS) and optical coherence tomography (OCT) are meanwhile established tools for detailed assessment of the coronary arteries. Their use might further increase with automated tissue characterization, three dimensional reconstruction, integration in angiography systems, and new emerging techniques. Parameters of fluid tissue interactions are important factors in the pathogenesis of atherosclerosis. These parameters can be calculated using computational fluid dynamics based on three dimensional models of the coronary vessels which can be derived from various sources including multislice computed tomography (MSCT), C-Arm CT or 3D reconstructed IVUS or OCT. Their use in the clinical setting has yet to be determined especially with regard to their ability in increasing treatment efficiency and clinical outcome.

Impact of intra-arterial injection parameters on arterial, capillary, and venous time-concentration curves in a canine model.

BACKGROUND AND PURPOSE: Recent advances in flat panel detector angiographic equipment have provided the opportunity to obtain physiologic and anatomic information from angiographic examinations. To exploit this possibility, one must understand the factors that affect the bolus geometry of an intra-arterial injection of contrast medium. It was our purpose to examine these factors in a canine model. MATERIALS AND METHODS: Under an institutionally approved protocol conforming to Guide for the Care and Use of Laboratory Animals of the National Institutes of Health, 7 canines were placed under general anesthesia with isoflurane and propofol. Through a 5F catheter placed into the right common carotid artery, a series of biplane angiographic acquisitions was obtained to examine the effects caused by variation in the volume of injection, the rate of injection, the duration of injection, the concentration of contrast medium, and the catheter position on arterial, capillary, and venous opacification. The results of each injection protocol were determined from analysis of a time-contrast concentration curve derived from locations over an artery, in brain parenchyma, and over a vein. The curve was generated from 2D digital subtraction angiography acquisitions by using prototype software. The area under the curve, the amplitude of the curve, and the time to peak (TTP) were analyzed separately for each injection parameter. RESULTS: Changes in the injection protocols resulted in predictable changes in the time-concentration curves. The injection parameter that contributed most to maximum opacification was the volume of contrast medium injected. When the injection rate was fixed and the volume was varied, there was an increase in opacification (maximal) proportional to the injected volume. The injected volume also had an indirect (secondary) impact on the temporal characteristics of the opacification. The time-concentration curve became wider, and the peak was shifted to the right as the injection duration increased. The impact of injected volume on maximal opacification was significant (P < .0001), regardless of the site of measurement (artery, tissue, and vein); however, the impact on the temporal characteristics of the time-concentration curve reached statistical significance only in measurements made in the artery and the vein (P < .05), but not in the tissue (P > .1). The impact of injected volume on maximal opacification became nonproportional in the tissue and vein when the volume was very large (>12 mL). Increasing the concentration of contrast medium resulted in a nonproportional increase in the height of the time-concentration curves (P < .05). Injection rate had an impact on both maximal opacification and TTP. The impact on TTP occurred only when the injection rate was very slow (1 mL/s). Changes of concentration had a similar impact on the time-concentration curve. Catheter position did not cause significant alterations in the shape of the curves. CONCLUSIONS: There were predictable effects from modification of injection parameters on the contrast bolus geometry and on time-concentration curves as measured in an artery, brain parenchyma, or a vein. The amplitude, TTP, and area under the time-concentration curve depend mainly and proportionally on the amount of iodine traversing the vasculature per second. Other injection parameters were of less importance in defining bolus geometry. These findings mimic those observed in studies of parameters affecting bolus geometry following an intravenous injection.

C-arm CT measurement of cerebral blood volume: an experimental study in canines.

BACKGROUND AND PURPOSE: Cerebral blood volume (CBV) is an important parameter in estimating the viability of brain tissue following an ischemic event. We tested the hypothesis that C-arm CT measurements of CBV would correlate well with those made with perfusion CT (PCT). MATERIALS AND METHODS: CBV was measured in 12 canines by using PCT and C-arm CT. Two measurements with each technique were made on each animal; a different injection protocol was used for each of these techniques. PCT was performed by using a 64-section V-scanner. C-arm CT was performed by using a biplane Artis dBA system. PCT images were transferred to a commercially available workstation for postprocessing and analysis; C-arm CT images were transferred to a commercially available workstation for postprocessing and analysis by using prototype software. From each animal, 2 sections from each technique were selected for analysis. RESULTS: There was good agreement of both the color maps and absolute numbers between the 2 techniques. The maximum and mean deviations of values between the 2 techniques for the first 5 animals were 30.20% and 7.82%; for the second 7 animals, these values were 26.79% and 7.40%. The maximum and mean deviations between the 2 C-arm CT studies performed on the first 5 animals were 33.15% and 12.24%; for the second 7 animals, these values were 41.15% and 10.89%. CONCLUSIONS: In these healthy animals, measurement of CBV with C-arm CT compared well with measurements made with PCT.

[Intravascular optical coherence tomography: differentiation of atherosclerotic plaques and quantification of vessel dimensions in crural arterial specimens]. / Intravaskuläre optische Kohärenztomographie: Unterscheidung verschiedener Plaquetypen und Vermessung von Gefässdimensionen in atherosklerotischen Unterschenkelarterien.

PURPOSE: Intravascular optical coherence tomography (OCT) is a new technique based on infrared light that visualizes the arteries with a resolution of 10-20 microm. Intravascular ultrasound (IVUS) is the current in vivo reference standard and provides a resolution of 100-150 microm. This study compared OCT to IVUS and histopathology with respect to the ability to differentiate atherosclerotic plaques and quantify vascular dimensions in peripheral crural arteries ex vivo. MATERIALS AND METHODS: 50 segments of atherosclerotic arteries derived from five amputated human lower extremities were examined. The different plaque types (fibrous, high-lipid content, calcified) were assigned by two independent examiners, and the sensitivity and specificity of OCT in comparison with histopathology as well as intra- and interobserver consensus were calculated. A comparison of OCT with IVUS addressed the parameters: luminal area (LA), vascular wall area (VA) and plaque area (PA). RESULTS: When comparing OCT and histopathology with respect to the differentiation of various plaque types, sensitivities of 81 % and specificities of 89 % for fibrous plaques, of 100 % and 93 % for lipid-rich plaques and of 80 % and 89 % for calcified plaques were achieved (overall correlation 83 %). Intra- and interobserver consensus was very high (kappa = 0.86 and kappa = 0.89, p < 0.001, respectively). There was also a high correlation between quantitative measurements (Bland-Altman plot [LA]: mean bias, 0.1 mm(2) accuracy +/- 1.8 mm(2), r = 0.95 [p < 0.001] Bland-Altman plot [VA]: mean bias, 0.3 mm(2) accuracy +/- 2.3 mm(2), r = 0.94 [p < 0.001] Bland-Altman plot [PA]: mean bias, 0.4 mm(2) accuracy +/- 2.3 mm(2), r = 0.80 [p < 0.01]. CONCLUSION: OCT allows the differentiation of atherosclerotic plaque types in crural arteries with high accuracy compared to histopathology. Quantitative measurements show a high correlation with IVUS, the current reference standard.