Experimental study using phantom models of cerebral aneurysms and 4D-DSA to measure blood flow on 3D-color-coded images.

BACKGROUND: The current 3D-iFlow application can only measure the arrival time of contrast media through intensity values. If the flow rate could be estimated by 3D-iFlow, patient-specific hemodynamics could be determined within the scope of normal diagnostic management, eliminating the need for additional resources for blood flow rate estimation. OBJECTIVE: The aim of this study is to develop and validate a method for measuring the flow rate by data obtained from 3D-iFlow images - a prototype application in Four-dimensional digital subtraction angiography (4D-DSA). METHODS: Using phantom model and experimental circuit with circulating glycerin solution, an equation for the relationship between contrast media intensity and flow rate was developed. Applying the equation to the aneurysm phantom models, the derived flow rate was evaluated. RESULTS: The average errors between the derived flow rate and setting flow rate became larger when the glycerin flow and the X-rays from the X-ray tube of the angiography system were parallel to each other or when the measurement point included overlaps with other contrast enhanced areas. CONCLUSION: Although the error increases dependent on the imaging direction and overlap of contrast enhanced area, the developed equation can estimate the flow rate using the image intensity value measured on 3D-iFlow based on 4D-DSA.

Hemodynamic and morphological differences in cerebral aneurysms between before and after rupture.

OBJECTIVE: Although it has been proposed that aneurysm morphology is different after rupture, detailed research of the morphological changes using 3D imaging acquired before and after rupture has not been conducted because of the difficulty of data collection. Similarly, hemodynamic changes due to morphological alterations after rupture have not been analyzed. The aim of this study was to investigate the changes in morphology and hemodynamics observed after aneurysm rupture. METHODS: For 21 cerebral aneurysms (21 patients) that ruptured during observation, 3D geometry of the aneurysms and parent arteries were reconstructed based on the angiographic images before and after their rupture. In addition, using the reconstructed geometry, blood flow was simulated by computational fluid dynamics (CFD) analysis. Morphological and hemodynamic parameters were calculated both before and after rupture, and their changes from before to after were compared. RESULTS: In the morphological parameters, statistically significantly higher values were observed after rupture in height (before: 5.5 ± 2.1 mm, after: 6.1 ± 2.0 mm; p < 0.0001), aspect ratio (p = 0.002), aneurysm volume (p = 0.04), and undulation index (p = 0.005). In terms of hemodynamic changes, the mean normalized wall shear stress (NWSS) decreased significantly (before: 5.4 × 10-1 ± 2.9 × 10-1, after: 4.4 × 10-1 ± 2.8 × 10-1; p < 0.001) as well as the other NWSS parameters, including maximum and minimum NWSS, which were associated with stagnant flow due to the morphological changes after rupture. CONCLUSIONS: Aneurysm morphology was found to change after rupture into an elongated and irregular geometry, accompanied by an increase in aneurysm volume. These morphological changes were also associated with statistically significant hemodynamic alterations that produced low wall sheer stress by stagnant flow. The authors' results also provide the opportunity to explore and develop a risk evaluation method for aneurysm rupture based on prerupture morphology and hemodynamics by further exploration in this direction.

Development of Machine Learning Model for Selecting the 1st Coil in the Treatment of Cerebral Aneurysms by Coil Embolization.

To achieve good treatment outcomes in coil embolization for cerebral aneurysms, it is important to select an appropriate 1st coil for each aneurysm since it serves as a frame to support the subsequent coils to be deployed. However, its selection as appropriate size and length from a wide variety of lineups is not easy, especially for inexperienced neurosurgeons. We developed a machine learning model (MLM) to predict the optimal size and length of the 1st coil by learning information on patients and aneurysms that were previously treated with coil embolization successfully. The accuracy rates of the MLM for the test data were 86.3% and 83.4% in the prediction of size and length, respectively. In addition, the accuracy rates for the 30 cases showed good prediction by the MLM when compared with two different skilled neurosurgeons. Although the accuracy rate of the well-experienced neurosurgeon is similar to MLM, the inexperienced neurosurgeon showed a worse rate and can benefit from the method.Clinical Relevance- The developed MLM has the potential to assist in the selection of the 1st coil for aneurysms. A technically and cost efficient supply chain in the treatment of aneurysms may also be achieved by MLM application.

Effects of different stent wire mesh densities on hemodynamics in aneurysms of different sizes.

BACKGROUND: Intracranial stents are used to treat aneurysms by diverting the blood flow from entering into the aneurysmal dome. Although delayed rupture is rare, clinical outcomes are extremely poor in such cases. Hemodynamics after stent deployment may be related to delayed rupture and a better understanding of the basic characteristics of pressure changes resulting from stent deployment is needed; therefore, this study investigated the relationships between hemodynamics in aneurysms of different sizes treated using stents of different wire mesh densities. METHODS: Using computational fluid dynamics analysis, parameters related to velocity, volume flow rate, pressure, and residual volume inside the aneurysm were evaluated in digital models of 5 basic aneurysms of differing sizes (Small, Medium, Medium-Large, Large, and Giant) and using 6 different types of stent (varying number of wires, stent pitch and wire mesh density) for each aneurysm. RESULTS: Regardless of the aneurysm size, the velocity inside the aneurysm and the volume flow rate into the aneurysm were observed to continuously decrease up to 89.2% and 78.1%, respectively, with increasing stent mesh density. In terms of pressure, for giant aneurysms, the pressure on the aneurysmal surface elevated to 10.3%, then decreased to 5.1% with increasing stent mesh density. However, in smaller aneurysms, this pressure continuously decreased with increasing stent mesh density. The flow-diverting effect of the stents was limited when a stent with low mesh density (under 20%) was used with a giant aneurysm. CONCLUSIONS: The present results indicate that the selection of appropriate stents according to aneurysm size may contribute to reduced risks of hemodynamic alternations related to stent deployment, which could reduce the incidence of delayed rupture.

Extraction of patient-specific boundary conditions from 4D-DSA and their influence on CFD simulations of cerebral aneurysms.

We developed a new technique for extracting patient-specific inflow conditions, such as the pulse cycle duration and blood flow velocity, from four-dimensional digital subtraction angiography images and experimentally examined its validity. The maximum error between the values extracted by the technique and measured values was 14.3%. We performed blood flow simulations and calculated representative haemodynamic parameters. The maximum differences between the parameters obtained using general and patient-specific inflow conditions were approximately 400%, 150%, and 50% for the velocity, normalised wall shear stress, and pressure loss coefficient, respectively. These results indicate that patient-specific conditions are critical for accurately reproducing aneurysmal haemodynamics.

Role of patient-specific blood properties in computational fluid dynamics simulation of flow diverter deployed cerebral aneurysms.

BACKGROUND: Hemodynamics and their clinical outcome of cerebral aneurysms treated with flow diverter (FD) stents have thus far been investigated using computational fluid dynamics (CFD) simulations. Although human blood is characterized as a non-Newtonian patientspecific fluid, non-patient-specific blood properties (PSBP) were applied in most extant studies. OBJECTIVE: To investigate the hemodynamic effects caused by PSBPs in aneurysms treated with FD stents. METHODS: We measured blood properties (density and viscosity) for 12 patients who underwent FD stent deployment. We conducted CFD simulations with the measured PSBPs and non-PSBPs quoted from previous studies. The average blood flow velocity and wall shear stress within the aneurysms were calculated and two simulation patterns were compared. RESULTS: The velocity and wall shear stress changed by 2.93% and 3.16% on average, respectively, without an FD stent deployed. Conversely, with the FD stents deployed, the change rates increased to 11.1% and 9.06% on average, respectively. CONCLUSIONS: The change in hemodynamic parameters if PSBPs are considered, may not be negligible when conducting CFD simulations of FD stent deployed aneurysms To obtain an adequate hemodynamic environment for cerebral aneurysms with FD stents deployed, it is recommended to use PSBPs for CFD simulations.

Hemodynamic Characteristics and Clinical Outcome for Intracranial Aneurysms Treated with the Derivo Embolization Device, a Novel Second-Generation Flow Diverter.

BACKGROUND: We investigated the relationship between hemodynamic characteristics and clinical outcomes for aneurysms treated by the Derivo embolization device, a novel second-generation flow-diverter stent, using computational fluid dynamics (CFD). METHODS: Data were retrospectively obtained from 2 centers between 2017 and 2019. During the period, 23 patients were treated for 23 aneurysms with the Derivo embolization device. In 17 patients we were able to conduct CFD analysis as 6 were excluded due to precoiling, unsuitable arterial geometry, and complex geometric form. Aneurysm occlusion was rated with the O'Kelly-Marotta grading scale on digital subtraction angiography 6 months after stent placement in all patients. Hemodynamic and morphologic parameters were statistically compared between 2 groups: with full occlusion and with a remnant. RESULTS: Full occlusion was observed in 17 of 23 (73.9%) patients. In the group suitable for CFD analysis, we observed 13 fully occluded aneurysms and 4 with any remnant (specifically 1 O'Kelly-Marotta C, 1 B, and 2 A). The energy loss per volume, which indicates the energy loss through the aneurysm, was significantly larger in prestenting and post stenting (P < 0.05) in the complete occlusion cases. In addition, the inflow concentration index and inflow area ratio of the remnant cases were significantly larger and lower, respectively (P < 0.05). CONCLUSIONS: Our CFD results indicate that the energy loss involved with the blood flow passing through an aneurysm and concentrated inflow into an aneurysm were the most important factors to determine whether an aneurysm will become a complete occlusion or remnant case.

Development of a Virtual Stent Deployment Application to Estimate Patient-Specific Braided Stent Sizes.

A virtual stent deployment application was developed to estimate the appropriate and patient-specific size of a braided stent for patients who undergo endovascular treatment for intracranial aneurysms. Comparing between the simulated deployed and the actual stents, we evaluated the accuracy of the simulation results. Our results indicated that lengths of the virtual and actual stents matched well despite the actual stent being affected by a geometrical change of the parent artery.Clinical Relevance-Surgeons need to be well-experienced to select an appropriate braided stent size for endovascular treatment of intracranial aneurysms, because the actual length of the deployed stent changes. This simulation will be helpful to make tailor-made surgical planning regardless of the surgeons' individual skill level.

Computational fluid dynamic analysis of the initiation of cerebral aneurysms.

OBJECTIVE: Relationships between aneurysm initiation and hemodynamic factors remain unclear since de novo aneurysms are rarely observed. Most previous computational fluid dynamics (CFD) studies have used artificially reproduced vessel geometries before aneurysm initiation for analysis. In this study, the authors investigated the hemodynamic factors related to aneurysm initiation by using angiographic images in patients with cerebral aneurysms taken before and after an aneurysm formation. METHODS: The authors identified 10 cases of de novo aneurysms in patients who underwent follow-up examinations for existing cerebral aneurysms located at a different vessel. The authors then reconstructed the vessel geometry from the images that were taken before aneurysm initiation. In addition, 34 arterial locations without aneurysms were selected as control cases. Hemodynamic parameters acting on the arterial walls were calculated by CFD analysis. RESULTS: In all de novo cases, the aneurysmal initiation area corresponded to the highest wall shear stress divergence (WSSD point), which indicated that there was a strong tensile force on the arterial wall at the initiation area. The other previously reported parameters did not show such correlations. Additionally, the pressure loss coefficient (PLc) was statistically significantly higher in the de novo cases (p < 0.01). The blood flow impact on the bifurcation apex, or the secondary flow accompanied by vortices, resulted in high tensile forces and high total pressure loss acting on the vessel wall. CONCLUSIONS: Aneurysm initiation may be more likely in an area where both tensile forces acting on the vessel wall and total pressure loss are large.

Hemodynamic Investigation of the Effectiveness of a Two Overlapping Flow Diverter Configuration for Cerebral Aneurysm Treatment.

Flow diverters (FDs) are widely employed as endovascular treatment devices for large or wide-neck cerebral aneurysms. Occasionally, overlapped FDs are deployed to enhance the flow diversion effect. In this study, we investigated the hemodynamics of overlapping FDs via computational fluid dynamics (CFD) simulations. We reproduced the arterial geometry of a patient who had experienced the deployment of two overlapping FDs. We utilized two stent patterns, namely the patterns for one FD and two overlapping FDs. We calculated the velocity, mass flow rate, wall shear stress, and pressure loss coefficient as well as their change rates for each pattern relative to the no-FD pattern results. The CFD simulation results indicated that the characteristics of the blood flow inside the aneurysm were minimally affected by the deployment of a single FD; in contrast, the overlapping FD pattern results revealed significant changes in the flow. Further, the velocity at an inspection plane within the aneurysm sac decreased by up to 92.2% and 31.0% in the cases of the overlapping and single FD patterns, respectively, relative to the no-FD pattern. The simulations successfully reproduced the hemodynamics, and the qualitative and quantitative investigations are meaningful with regard to the clinical outcomes of overlapped FD deployment.

Hemodynamic and Morphologic Factors Related to Coil Compaction in Basilar Artery Tip Aneurysms.

OBJECTIVE: Coil compaction is directly related to the degree of cerebral aneurysmal recanalization. The degree of recanalization (DoR) was quantified by measuring the volume vacated by coil deformation. The purpose of this study was to clarify the hemodynamic and morphologic factors associated with coil compaction. METHODS: Computational fluid dynamics simulations were performed on 28 middle-size (5-10 mm) unruptured basilar artery tip aneurysms. The DoR was measured by comparing the coil mass shape obtained from three-dimensional digital subtraction angiography data immediately after coil embolization and again within 1-2 years of follow-up. Deployed coils were modeled using a virtual coiling technique for computational fluid dynamics simulations. Hemodynamic and morphologic factors to predict the DoR were derived using multiple linear regression. RESULTS: Aneurysmal neck area, the maximum pressure generated on the neck surface after coil embolization, and the high-pressure position on the neck surface predicted DoR with statistic significance (P < 0.001, P < 0.001, P = 0.004, respectively). The DoR tended to increase when the neck area was large, the pressure generated on the coils was high, and the high-pressure position was close to the center of the neck surface. The volume embolization ratio was not statistically relevant for the DoR in the cases of this study. CONCLUSIONS: Coil compaction occurs in cerebral aneurysms with a wide neck, high pressure generated on the coils, and high pressure in the center of the neck surface. Establishing the DoR can contribute to the prediction of recanalization after coil embolization.

Blood Flow Analysis in Coil Embolized Aneurysms: Difference between Porous Media and Real Coil Geometry Model.

To clarify the mechanism of aneurysmal recanalization, it is necessary to understand the characteristics of the blood flow inside the aneurysm in particular the flow resistance generated by the coil. In studies using computational fluid dynamics (CFD), mainly two approaches have been used to model the coil embolized aneurysm; modeling the coils as porous media or by real coil geometries. In this study, we calculated the pressure drop along a vessel through a coiled region modeled as porous media or by real coil geometry and compared the pressure drop generated by the two coil models. The porous media model was described by Darcy's law and Ergun's equation, while the real coil geometry was generated using finite element method (FEM) structural analysis. We calculated the pressure drop for inlet velocities from 0.1 m/s to 1.0 m/s in steps of 0.1 m/s. Our results indicated that the porous media model may produce larger pressure drops than the real coil geometry model under low packing density. The value of the pressure drop was also changed due to the difference of coil distribution even if the packing density was the same.

Hemodynamic Change In A Cerebral Aneurysm Treated By Double Stenting Technique.

Rupture of cerebral aneurysms often causes subarachnoid hemorrhage which is a life-threatening condition with high mortality rates. Larger aneurysms are believed to be more likely to rupture and should therefore be treated. Recently, flow diverters (FDs) are widely used to treat large or wide neck aneurysms. However, it can be difficult to treat them by deployment of a single FD because of its insufficient flow disturbance. To overcome this problem, double stenting technique is sometimes applied with the aim to improve the effect of blood velocity reduction. In this study, we used computational fluid dynamics (CFD) to investigate the hemodynamic changes in an aneurysm when deploying virtual FDs. The results showed that the characteristics of the blood flow field inside the aneurysm did not changed much after the deployment of a single FD but underwent a large change after the deployment of two FDs. Furthermore, the velocity reduction in the aneurysm sac at a plane away from the parent artery increased from 25.9% to 92.8% when two FDs were deployed instead of one compared to no stenting. Double stenting was effective to decrease blood velocity in large or wide neck aneurysms.

Relationship between hemodynamic parameters and cerebral aneurysm initiation.

Research on the relationship between cerebralaneurysm initiation and hemodynamic parameters, but several open questions remain on initiation and growth mechanisms of cerebral aneurysms. If factors contributing to initiation were identified, it would be possible to predict the initiation of aneurysms. The purpose of this study is to investigate the relationship between cerebral aneurysm initiation and hemodynamic factors. Blood flow simulations in aneurysms of three patients were performed using computational fluid dynamics (CFD) based on the cerebral blood vessel geometry before aneurysm initiation. We evaluated pressure, wall shear stress (WSS), wall shear stress gradient (WSSG), oscillatory shear index (OSI) and gradient oscillatory number (GON) since these factors are known to be associated with aneurysmal initiation. We also focused on the wall shear stress divergence (WSSD) in particular on the direction of WSS. Our results indicated that only high WSSD regions corresponded to the initiation regions, and the value of WSSD was remarkably high. Stretching force to the vessel wall may be related to the initiation of cerebral aneurysms.

Computational Experiments on the Step and Frequency Responses of a Three-Axis Thermal Accelerometer.

The sensor response has been reported to become highly nonlinear when the acceleration added to a thermal accelerator is very large, so the same response can be observed for two accelerations with different magnitudes and opposite signs. Some papers have reported the frequency response for the horizontal acceleration to be a first-order system, while others have reported it to be a second-order system. The response for the vertical acceleration has not been studied. In this study, computational experiments were performed to examine the step and frequency responses of a three-axis thermal accelerometer. The results showed that monitoring the temperatures at two positions and making use of cross-axis sensitivity allow a unique acceleration to be determined even when the range of the vertical acceleration is very large (e.g., -10,000-10,000 g). The frequency response was proven to be a second-order system for horizontal acceleration and a third-order system for vertical acceleration.