Agreement between single plane and biplane derived angiographic fractional flow reserve in patients with intermediate coronary artery stenosis.

Fractional flow reserve (FFR) is often used to evaluate the physiological severity of intermediate coronary stenoses, but less-invasive assessment methods are desirable. We evaluated the feasibility of angiographic FFR (angioFFR) calculated from two projections acquired simultaneously by a biplane C-arm system and angioFFR calculated from two projections acquired independently by one plane of the same biplane C-arm system. AngioFFR was validated against FFR in terms of detection of hemodynamically relevant coronary artery stenoses. Twenty-two Patients who underwent angiography and FFR for coronary artery disease were included. We used a non-commercial prototype to calculate biplane angioFFR for 22 vessels (19 LAD, 1 LCx, 2 RCA) and single plane angioFFR for 17 of the same 22 vessels. FFR < 0.8 was measured in 8 vessels. The Pearson correlation coefficients with FFR were 0.55 for single plane angioFFR and 0.61 for biplane angioFFR and the diagnostic accuracies were 88% (95% CI 73-100%) for single plane angioFFR and 86% (95% CI 72-100%) for biplane angioFFR. Bland-Altman plots revealed that compared with FFR, the limits of agreement for single plane angioFFR were - 0.07 to 0.19 (mean difference 0.06, p = 0.002) and the limits of agreement for biplane FFR were - 0.09 to 0.15 (mean difference 0.03, p = 0.03). In conclusion, angioFFR calculated from single or biplane acquisitions by a biplane C-arm is feasible and may evolve to a tool for less invasive imaging-based assessment of myocardial ischemia.

An Automated Workflow for Hemodynamic Computations in Cerebral Aneurysms.

In recent years, computational fluid dynamics (CFD) has become a valuable tool for investigating hemodynamics in cerebral aneurysms. CFD provides flow-related quantities, which have been shown to have a potential impact on aneurysm growth and risk of rupture. However, the adoption of CFD tools in clinical settings is currently limited by the high computational cost and the engineering expertise required for employing these tools, e.g., for mesh generation, appropriate choice of spatial and temporal resolution, and of boundary conditions. Herein, we address these challenges by introducing a practical and robust methodology, focusing on computational performance and minimizing user interaction through automated parameter selection. We propose a fully automated pipeline that covers the steps from a patient-specific anatomical model to results, based on a fast, graphics processing unit- (GPU-) accelerated CFD solver and a parameter selection methodology. We use a reduced order model to compute the initial estimates of the spatial and temporal resolutions and an iterative approach that further adjusts the resolution during the simulation without user interaction. The pipeline and the solver are validated based on previously published results, and by comparing the results obtained for 20 cerebral aneurysm cases with those generated by a state-of-the-art commercial solver (Ansys CFX, Canonsburg PA). The automatically selected spatial and temporal resolutions lead to results which closely agree with the state-of-the-art, with an average relative difference of only 2%. Due to the GPU-based parallelization, simulations are computationally efficient, with a median computation time of 40 minutes per simulation.

Deep neural networks for ECG-free cardiac phase and end-diastolic frame detection on coronary angiographies.

Invasive coronary angiography (ICA) is the gold standard in Coronary Artery Disease (CAD) imaging. Detection of the end-diastolic frame (EDF) and, in general, cardiac phase detection on each temporal frame of a coronary angiography acquisition is of significant importance for the anatomical and non-invasive functional assessment of CAD. This task is generally performed via manual frame selection or semi-automated selection based on simultaneously acquired ECG signals - thus introducing the requirement of simultaneous ECG recordings. In this paper, we evaluate the performance of a purely image based workflow relying on deep neural networks for fully automated cardiac phase and EDF detection on coronary angiographies. A first deep neural network (DNN), trained to detect coronary arteries, is employed to preselect a subset of frames in which coronary arteries are well visible. A second DNN predicts cardiac phase labels for each frame. Only in the training and evaluation phases for the second DNN, ECG signals are used to provide ground truth labels for each angiographic frame. The networks were trained on 56,655 coronary angiographies from 6820 patients and evaluated on 20,780 coronary angiographies from 6261 patients. No exclusion criteria related to patient state (stable or acute CAD), previous interventions (PCI or CABG), or pathology were formulated. Cardiac phase detection had an accuracy of 98.8 %, a sensitivity of 99.3 % and a specificity of 97.6 % on the evaluation set. EDF prediction had a precision of 98.4 % and a recall of 97.9 %. Several sub-group analyses were performed, indicating that the cardiac phase detection performance is largely independent from acquisition angles, the heart rate of the patient, and the angiographic view (LCA / RCA). The average execution time of cardiac phase detection for one angiographic series was on average less than five seconds on a standard workstation. We conclude that the proposed image based workflow potentially obviates the need for manual frame selection and ECG acquisition, representing a relevant step towards automated CAD assessment.

Rupture Risk of Small Unruptured Intracranial Aneurysms in Japanese Adults.

Background and Purpose- Therapeutic decision making for small unruptured intracranial aneurysms (<10 mm) is difficult. We aimed to develop a rupture risk model for small intracranial aneurysms in Japanese adults, including clinical, morphological, and hemodynamic parameters. Methods- We analyzed 338 small unruptured aneurysms; 35 ruptured during the observation period, and 303 remained stable. Clinical, morphological, and hemodynamic parameters were considered. Computational fluid dynamics was used to calculate hemodynamic parameters based on computed tomography images of all aneurysms in their unruptured state. Differences between the ruptured and unruptured groups were tested by the Mann-Whitney U or Fisher exact tests. Multivariate logistic regression was applied to obtain a rupture risk model. Its predictive ability was investigated by receiver operating characteristic analysis. Results- The risk model revealed that rupture may be more likely to in younger patients (odds ratio [OR], 0.92 for each age increase of 1 year [95% CI, 0.88-0.96] P<0.001) with multiple aneurysms (OR, 2.58 [95% CI, 1.07-6.19] P=0.03), located at a bifurcation (OR, 5.45 [95% CI, 1.87-15.85] P=0.002), with a bleb (OR, 4.09 [95% CI, 1.42-11.79] P=0.009), larger length (OR, 1.91 for each increase of 1 mm [95% CI, 1.42-2.57] P<0.001), and lower pressure loss coefficient (OR, 0.33 for each decrease of 1 unit [95% CI, 0.14-0.77] P=0.01). The sensitivity, specificity, and area under the curve were 0.800, 0.752, and 0.826 (95% CI, 0.739-0.914) respectively. Conclusions- Younger age, presence of multiple aneurysms, location at a bifurcation, presence of a bleb, larger length, and lower pressure loss coefficient were identified as risk factors for rupture of small intracranial aneurysms. The risk model should be validated in further studies.

Viewpoint planning for quantitative coronary angiography.

PURPOSE: In coronary angiography, the condition of myocardial blood supply is assessed by analyzing 2-D X-ray projections of contrasted coronary arteries. This is done using a flexible C-arm system. Due to the X-ray immanent dimensionality reduction projecting the 3-D scene onto a 2-D image, the viewpoint is critical to guarantee an appropriate view onto the affected artery and, thus, enable reliable diagnosis. In this work, we introduce an algorithm computing optimal viewpoints for the assessment of coronary arteries without the need for 3-D models. METHODS: We introduce the concept of optimal viewpoint planning solely based on a single angiographic X-ray image. The subsequent viewpoint is computed such that it is rotated precisely around a vessel, while minimizing foreshortening. RESULTS: Our algorithm reduces foreshortening substantially compared to the input view and completely eliminates it for [Formula: see text] rotations. Rotations around isocentered foreshortening-free vessels passing the isocenter are exact. The precision, however, decreases when the vessel is off-centered or foreshortened. We evaluate worst-case boundaries, providing insight in the maximal inaccuracies to be expected. This can be utilized to design viewpoints guaranteeing desired requirements, e.g., a true rotation around the vessel of at minimum [Formula: see text]. In addition, a phantom study is performed investigating the impact of input views to 3-D quantitative coronary angiography (QCA). CONCLUSION: We introduce an algorithm for optimal viewpoint planning from a single angiographic X-ray image. The quality of the second viewpoint-i.e., vessel foreshortening and true rotation around vessel-depends on the first viewpoint selected by the physician; however, our computed viewpoint is guaranteed to reduce the initial foreshortening. Our novel approach uses fluoroscopy images only and, thus, seamlessly integrates with the current clinical workflow for coronary assessment. In addition, it can be implemented in the QCA workflow without increasing user interaction, making vessel-shape reconstruction more stable by standardizing viewpoints.

Tomographic particle image velocimetry for the validation of hemodynamic simulations in an intracranial aneurysm.

Image-based blood flow simulations can provide detailed hemodynamic information in diseased vessels such as intracranial aneurysms. However, validation is essential to evaluate the accuracy of these computations and further improve their acceptance among physicians. In this regard, tomographic particle image velocimetry was used to measure the flow characteristics in a patient specific aneurysm phantom model. Additionally, computational fluid dynamics (CFD) simulations were carried out using a well accepted commercial software package and a clinical research prototype, respectively. The comparison between in-vitro measurement and in-silico computations reveals a good qualitative agreement. Further, computations based on classical CFD agreed well with results from a clinical research prototype. Hence, the results of this study demonstrate the usability of numerical methods to obtain realistic blood flow predictions in a clinical context.

Patient-individualized boundary conditions for CFD simulations using time-resolved 3D angiography.

PURPOSE: Hemodynamic simulations are of increasing interest for the assessment of aneurysmal rupture risk and treatment planning. Achievement of accurate simulation results requires the usage of several patient-individual boundary conditions, such as a geometric model of the vasculature but also individualized inflow conditions. METHODS: We propose the automatic estimation of various parameters for boundary conditions for computational fluid dynamics (CFD) based on a single 3D rotational angiography scan, also showing contrast agent inflow. First the data are reconstructed, and a patient-specific vessel model can be generated in the usual way. For this work, we optimize the inflow waveform based on two parameters, the mean velocity and pulsatility. We use statistical analysis of the measurable velocity distribution in the vessel segment to estimate the mean velocity. An iterative optimization scheme based on CFD and virtual angiography is utilized to estimate the inflow pulsatility. Furthermore, we present methods to automatically determine the heart rate and synchronize the inflow waveform to the patient's heart beat, based on time-intensity curves extracted from the rotational angiogram. This will result in a patient-individualized inflow velocity curve. RESULTS: The proposed methods were evaluated on two clinical datasets. Based on the vascular geometries, synthetic rotational angiography data was generated to allow a quantitative validation of our approach against ground truth data. We observed an average error of approximately [Formula: see text] for the mean velocity, [Formula: see text] for the pulsatility. The heart rate was estimated very precisely with an average error of about [Formula: see text], which corresponds to about 6 ms error for the duration of one cardiac cycle. Furthermore, a qualitative comparison of measured time-intensity curves from the real data and patient-specific simulated ones shows an excellent match. CONCLUSION: The presented methods have the potential to accurately estimate patient-specific boundary conditions from a single dedicated rotational scan.

Comparison of Fractional Flow Reserve Based on Computational Fluid Dynamics Modeling Using Coronary Angiographic Vessel Morphology Versus Invasively Measured Fractional Flow Reserve.

Invasive fractional flow reserve (FFRinvasive), although gold standard to identify hemodynamically relevant coronary stenoses, is time consuming and potentially associated with complications. We developed and evaluated a new approach to determine lesion-specific FFR on the basis of coronary anatomy as visualized by invasive coronary angiography (FFRangio): 100 coronary lesions (50% to 90% diameter stenosis) in 73 patients (48 men, 25 women; mean age 67 ± 9 years) were studied. On the basis of coronary angiograms acquired at rest from 2 views at angulations at least 30° apart, a PC-based computational fluid dynamics modeling software used personalized boundary conditions determined from 3-dimensional reconstructed angiography, heart rate, and blood pressure to derive FFRangio. The results were compared with FFRinvasive. Interobserver variability was determined in a subset of 25 narrowings. Twenty-nine of 100 coronary lesions were hemodynamically significant (FFRinvasive ≤ 0.80). FFRangio identified these with an accuracy of 90%, sensitivity of 79%, specificity of 94%, positive predictive value of 85%, and negative predictive value of 92%. The area under the receiver operating characteristic curve was 0.93. Correlation between FFRinvasive (mean: 0.84 ± 0.11) and FFRangio (mean: 0.85 ± 0.12) was r = 0.85. Interobserver variability of FFRangio was low, with a correlation of r = 0.88. In conclusion, estimation of coronary FFR with PC-based computational fluid dynamics modeling on the basis of lesion morphology as determined by invasive angiography is possible with high diagnostic accuracy compared to invasive measurements.

Verification of a research prototype for hemodynamic analysis of cerebral aneurysms.

Owing to its clinical importance, there has been a growing body of research on understanding the hemodynamics of cerebral aneurysms. Traditionally, this work has been performed using general-purpose, state-of-the-art commercial solvers. This has meant requiring engineering expertise for making appropriate choices on the geometric discretization, time-step selection, choice of boundary conditions etc. Recently, a CFD research prototype has been developed (Siemens Healthcare GmbH, Prototype - not for diagnostic use) for end-to-end analysis of aneurysm hemodynamics. This prototype enables anatomical model preparation, hemodynamic computations, advanced visualizations and quantitative analysis capabilities. In this study, we investigate the accuracy of the hemodynamic solver in the prototype against a commercially available CFD solver ANSYS CFX 16.0 (ANSYS Inc., Canonsburg, PA, www.ansys.com) retrospectively on a sample of twenty patient-derived aneurysm models, and show good agreement of hemodynamic parameters of interest.

Bringing hemodynamic simulations closer to the clinics: a CFD prototype study for intracranial aneurysms.

Computational Fluid Dynamics enables the investigation of patient-specific hemodynamics for rupture predictions and treatment support of intracranial aneurysms. However, due to numerous simplifications to decrease the computations effort, clinical applicability is limited until now. To overcome this situation a clinical research software prototype was tested that can be easily operated by attending physicians. In order to evaluate the accuracy of this prototype, four patient-specific intracranial aneurysms were investigated using four different spatial resolutions. The results demonstrate that physicians were able to generate hemodynamic predictions within several minutes at low spatial resolution. However, depending on the parameter of interest and the desired accuracy, higher resolutions are required, which will lead to an increase of computational times that still look very attractive towards clinical usability. The study shows that the next step towards an applicable individualized therapy for patients harboring intracranial aneurysms can be done. However, further in vivo validations are required to guarantee realistic predictions in future studies.

A fully-automatic locally adaptive thresholding algorithm for blood vessel segmentation in 3D digital subtraction angiography.

Subarachnoid hemorrhage due to a ruptured cerebral aneurysm is still a devastating disease. Planning of endovascular aneurysm therapy is increasingly based on hemodynamic simulations necessitating reliable vessel segmentation and accurate assessment of vessel diameters. In this work, we propose a fully-automatic, locally adaptive, gradient-based thresholding algorithm. Our approach consists of two steps. First, we estimate the parameters of a global thresholding algorithm using an iterative process. Then, a locally adaptive version of the approach is applied using the estimated parameters. We evaluated both methods on 8 clinical 3D DSA cases. Additionally, we propose a way to select a reference segmentation based on 2D DSA measurements. For large vessels such as the internal carotid artery, our results show very high sensitivity (97.4%), precision (98.7%) and Dice-coefficient (98.0%) with our reference segmentation. Similar results (sensitivity: 95.7%, precision: 88.9% and Dice-coefficient: 90.7%) are achieved for smaller vessels of approximately 1mm diameter.

A Fourier-based approach to the angiographic assessment of flow diverter efficacy in the treatment of cerebral aneurysms.

Flow diversion is an emerging endovascular treatment option for cerebral aneurysms. Quantitative assessment of hemodynamic changes induced by flow diversion can aid clinical decision making in the treatment of cerebral aneurysms. In this article, besides summarizing past key research efforts, we propose a novel metric for the angiographic assessment of flow diverter deployments in the treatment of cerebral aneurysms. By analyzing the frequency spectra of signals derived from digital subtraction angiography (DSA) series, the metric aims to quantify the prevalence of frequency components that correspond to the patient-specific heart rate. Indicating the decoupling of aneurysms from healthy blood circulation, our proposed metric could advance clinical guidelines for treatment success prediction. The very promising results of a retrospective feasibility study on 26 DSA series warrant future efforts to study the validity of the proposed metric within a clinical setting.

Hemodynamics at the ostium of cerebral aneurysms with relation to post-treatment changes by a virtual flow diverter: a computational fluid dynamics study.

Computational fluid dynamics (CFD) techniques have been refined for modeling the hemodynamics in cerebral aneurysms. Recent interest has focused on understanding hemodynamic changes by treatment with a flow diverter (FD), i.e. a stent with a dense metal mesh which is placed across the ostium to divert the majority of flow away from the aneurysm. Potential complications include remnant inflow jets but, more seriously, aneurysm hemorrhage. For optimization of treatment outcome, a better understanding of the effects caused by the FD would be beneficial. In particular, pressure and velocity distributions at the aneurysm ostium are of interest, as they will be directly affected by the FD which in turn will influence post-treatment hemodynamics inside the aneurysm. Here, we report the results of a CFD study investigating the relationship between pre-treatment and post-treatment velocities, pressures and wall shear stresses (WSS) in the aneurysm with corresponding hemodynamic conditions at the aneurysm ostium prior to treatment. The study was carried out using a dedicated CFD prototype which allows modeling the effects of a virtual FD integrated into patient-specific geometries utilizing Darcy's law. Velocities and WSS were reduced in all cases post FD treatment, pressure increased in one case. Heterogeneous distributions of the velocity magnitude were found at the ostium with focal maxima indicating potential risk zones for remnant inflow jets into the aneurysms. Pressures at the ostium correlated with pressure changes inside the aneurysm which could become a pre-treatment indicator for the evaluation of the suitability of a particular aneurysm for FD treatment.

A workflow for patient-individualized virtual angiogram generation based on CFD simulation.

Increasing interest is drawn on hemodynamic parameters for classifying the risk of rupture as well as treatment planning of cerebral aneurysms. A proposed method to obtain quantities such as wall shear stress, pressure, and blood flow velocity is to numerically simulate the blood flow using computational fluid dynamics (CFD) methods. For the validation of those calculated quantities, virtually generated angiograms, based on the CFD results, are increasingly used for a subsequent comparison with real, acquired angiograms. For the generation of virtual angiograms, several patient-specific parameters have to be incorporated to obtain virtual angiograms which match the acquired angiograms as best as possible. For this purpose, a workflow is presented and demonstrated involving multiple phantom and patient cases.

Tetrahedral vs. polyhedral mesh size evaluation on flow velocity and wall shear stress for cerebral hemodynamic simulation.

Haemodynamic factors, in particular wall shear stresses (WSSs) may have significant impact on growth and rupture of cerebral aneurysms. Without a means to measure WSS reliably in vivo, computational fluid dynamic (CFD) simulations are frequently employed to visualise and quantify blood flow from patient-specific computational models. With increasing interest in integrating these CFD simulations into pretreatment planning, a better understanding of the validity of the calculations in respect to computation parameters such as volume element type, mesh size and mesh composition is needed. In this study, CFD results for the two most common aneurysm types (saccular and terminal) are compared for polyhedral- vs. tetrahedral-based meshes and discussed regarding future clinical applications. For this purpose, a set of models were constructed for each aneurysm with spatially varying surface and volume mesh configurations (mesh size range: 5119-258, 481 volume elements). WSS distribution on the model wall and point-based velocity measurements were compared for each configuration model. Our results indicate a benefit of polyhedral meshes in respect to convergence speed and more homogeneous WSS patterns. Computational variations of WSS values and blood velocities are between 0.84 and 6.3% from the most simple mesh (tetrahedral elements only) and the most advanced mesh design investigated (polyhedral mesh with boundary layer).

Impact of tear location on hemodynamics in a type B aortic dissection investigated with computational fluid dynamics.

Stanford type B aortic dissections (TB-AD), which split the descending aorta in a true and false lumen, have better in-hospital survival than type A dissections affecting the ascending aorta. However, short-term and long-term prognosis for the individual patient remains challenging, with one in four patients not surviving after 3 years.

Tetrahedral and polyhedral mesh evaluation for cerebral hemodynamic simulation--a comparison.

Computational fluid dynamic (CFD) based on patient-specific medical imaging data has found widespread use for visualizing and quantifying hemodynamics in cerebrovascular disease such as cerebral aneurysms or stenotic vessels. This paper focuses on optimizing mesh parameters for CFD simulation of cerebral aneurysms. Valid blood flow simulations strongly depend on the mesh quality. Meshes with a coarse spatial resolution may lead to an inaccurate flow pattern. Meshes with a large number of elements will result in unnecessarily high computation time which is undesirable should CFD be used for planning in the interventional setting. Most CFD simulations reported for these vascular pathologies have used tetrahedral meshes. We illustrate the use of polyhedral volume elements in comparison to tetrahedral meshing on two different geometries, a sidewall aneurysm of the internal carotid artery and a basilar bifurcation aneurysm. The spatial mesh resolution ranges between 5,119 and 228,118 volume elements. The evaluation of the different meshes was based on the wall shear stress previously identified as a one possible parameter for assessing aneurysm growth. Polyhedral meshes showed better accuracy, lower memory demand, shorter computational speed and faster convergence behavior (on average 369 iterations less).

Intravascular optical coherence tomography: comparison with histopathology in atherosclerotic peripheral artery specimens.

PURPOSE: Intravascular optical coherence tomography (OCT) is a new imaging modality that provides microstructural information on atherosclerotic plaques and has an axial resolution of 10-20 microm. OCT of coronary arteries characterizes different atherosclerotic plaque components by their distinctive signal patterns. Peripheral human arteries were examined ex vivo by means of OCT, and attempts to distinguish among fibrous, lipid-rich, and calcified atherosclerotic plaques were made based on imaging criteria previously established for coronary arteries. MATERIALS AND METHODS: One hundred fifty-one atherosclerotic arterial segments were obtained from 15 below-knee amputations. OCT imaging criteria for different plaque types (fibrous, lipid-rich, calcified) were established in a subset of 30 arterial segments. The remaining 121 OCT images were analyzed by two independent readers. Each segment was divided into four quadrants. Agreement between histopathology and OCT was quantified by the kappa test of concordance, as were interobserver, intraobserver, and inter-method variability. RESULTS: Four hundred sixty-nine of 484 quadrants (97%) were available for comparison. Sensitivity and specificity for OCT criteria (consensus readers 1 and 2) were 86% and 86% for fibrous plaques, 78% and 93% for lipid-rich plaques, and 84% and 95% for calcified plaques, respectively (overall agreement, 84%). The interobserver and intraobserver reliabilities of OCT assessment were high (kappa values of 0.84 and 0.87, respectively). The inter-method agreement was 0.74 for consensus OCT versus consensus histology. CONCLUSIONS: OCT of peripheral human arteries ex vivo characterized different atherosclerotic plaque types with a high degree of agreement with histopathologic findings. Findings were comparable to those reported for coronary arteries. OCT promises to improve understanding of the progression or regression of peripheral atherosclerosis in vivo.