Radiomics-Based Predictive Nomogram for Assessing the Risk of Intracranial Aneurysms.

Aneurysm wall enhancement (AWE) has the potential to be used as an imaging biomarker for the risk stratification of intracranial aneurysms (IAs). Radiomics provides a refined approach to quantify and further characterize AWE's textural features. This study examines the performance of AWE quantification combined with clinical information in detecting symptomatic IAs. Ninety patients harboring 104 IAs (29 symptomatic and 75 asymptomatic) underwent high-resolution magnetic resonance imaging (HR-MRI). The assessment of AWE was performed using two different methods: 3D-AWE mapping and composite radiomics-based score (RadScore). The dataset was split into training and testing subsets. The testing set was used to build two different nomograms using each modality of AWE assessment combined with patients' clinical information and aneurysm morphological data. Finally, each nomogram was evaluated on an independent testing set. A total of 22 radiomic features were significantly different between symptomatic and asymptomatic IAs. The 3D-AWE mapping nomogram achieved an area under the curve (AUC) of 0.77 (63% accuracy, 78% sensitivity, and 58% specificity). The RadScore nomogram exhibited a better performance, achieving an AUC of 0.83 (77% accuracy, 89% sensitivity, and 73% specificity). The comprehensive analysis of IAs with the quantification of AWE data through radiomic analysis, patient clinical information, and morphological aneurysm metrics achieves a high accuracy in detecting symptomatic IA status.

Imaging of Intracranial Aneurysms: A Review of Standard and Advanced Imaging Techniques.

The treatment of intracranial aneurysms is dictated by its risk of rupture in the future. Several clinical and radiological risk factors for aneurysm rupture have been described and incorporated into prediction models. Despite the recent technological advancements in aneurysm imaging, linear length and visible irregularity with a bleb are the only radiological measure used in clinical prediction models. The purpose of this article is to summarize both the standard imaging techniques, including their limitations, and the advanced techniques being used experimentally to image aneurysms. It is expected that as our understanding of advanced techniques improves, and their ability to predict clinical events is demonstrated, they become an increasingly routine part of aneurysm assessment. It is important that neurovascular specialists understand the spectrum of imaging techniques available.

Radiomics-Based Predictive Nomogram for Assessing the Risk of Intracranial Aneurysms.

Background: Aneurysm wall enhancement (AWE) has the potential to be used as an imaging biomarker for the risk stratification of intracranial aneurysms (IAs). Radiomics provides a refined approach to quantify and further characterize AWE's textural features. This study examines the performance of AWE quantification combined with clinical information in detecting symptomatic IAs. Methods: Ninety patients harboring 104 IAs (29 symptomatic and 75 asymptomatic) underwent high-resolution magnetic resonance imaging (HR-MRI). The assessment of AWE was performed using two different methods: 3D-AWE mapping and composite radiomics-based score (RadScore). The dataset was split into training and testing subsets. The testing set was used to build two different nomograms using each modality of AWE assessment combined with patients' demographic information and aneurysm morphological data. Finally, each nomogram was evaluated on an independent testing set. Results: A total of 22 radiomic features were significantly different between symptomatic and asymptomatic IAs. The 3D-AWE Mapping nomogram achieved an area under the curve (AUC) of 0.77 (63% accuracy, 78% sensitivity and 58% specificity). The RadScore nomogram exhibited a better performance, achieving an AUC of 0.83 (77% accuracy, 89% sensitivity and 73% specificity). Conclusions: Combining AWE quantification through radiomic analysis with patient demographic data in a clinical nomogram achieved high accuracy in detecting symptomatic IAs.

Evaluating a 3D deep learning pipeline for cerebral vessel and intracranial aneurysm segmentation from computed tomography angiography-digital subtraction angiography image pairs.

OBJECTIVE: Computed tomography angiography (CTA) is the most widely used imaging modality for intracranial aneurysm (IA) management, yet it remains inferior to digital subtraction angiography (DSA) for IA detection, particularly of small IAs in the cavernous carotid region. The authors evaluated a deep learning pipeline for segmentation of vessels and IAs from CTA using coregistered, segmented DSA images as ground truth. METHODS: Using 50 paired CTA-DSA images, the authors trained (n = 27), validated (n = 3), and tested (n = 20) a deep learning model (3D DeepMedic) for cerebrovasculature segmentation from CTA. A landmark-based coregistration algorithm was used for registration and upsampling of CTA images to paired DSA images. Segmented vessels from the DSA were used as the ground truth. Accuracy of the model for vessel segmentation was evaluated using conventional metrics (dice similarity coefficient [DSC]) and vessel segmentation-specific metrics, like connectivity-area-length (CAL). On the test cases (20 IAs), 3 expert raters attempted to detect and segment IAs. For each rater, the authors recorded the rate of IA detection, and for detected IAs, raters segmented and calculated important IA morphology parameters to quantify the differences in IA segmentation by raters to segmentations by DeepMedic. The agreement between raters, DeepMedic, and ground truth was assessed using Krippendorf's alpha. RESULTS: In testing, the DeepMedic model yielded a CAL of 0.971 ± 0.007 and a DSC of 0.868 ± 0.008. The model prediction delineated all IAs and resulted in average error rates of < 10% for all IA morphometrics. Conversely, average IA detection accuracy by the raters was 0.653 (undetected IAs were present to a significantly greater degree on the ICA, likely due to those in the cavernous region, and were significantly smaller). Error rates for IA morphometrics in rater-segmented cases were significantly higher than in DeepMedic-segmented cases, particularly for neck (p = 0.003) and surface area (p = 0.04). For IA morphology, agreement between the raters was acceptable for most metrics, except for the undulation index (α = 0.36) and the nonsphericity index (α = 0.69). Agreement between DeepMedic and ground truth was consistently higher compared with that between expert raters and ground truth. CONCLUSIONS: This CTA segmentation network (DeepMedic trained on DSA-segmented vessels) provides a high-fidelity solution for CTA vessel segmentation, particularly for vessels and IAs in the carotid cavernous region.

Profiling of Circulating Gene Expression Reveals Molecular Signatures Associated with Intracranial Aneurysm Rupture Risk.

BACKGROUND: Following detection, rupture risk assessment for intracranial aneurysms (IAs) is critical. Towards molecular prognostics, we hypothesized that circulating blood RNA expression profiles are associated with IA risk. METHODS: We performed RNA sequencing on 68 blood samples from IA patients. Here, patients were categorized as either high or low risk by assessment of aneurysm size (≥ 5 mm = high risk) and Population, Hypertension, Age, Size, Earlier subarachnoid hemorrhage, Site (PHASES) score (≥ 1 = high risk). Modified F-statistics and Benjamini-Hochberg false discovery rate correction was performed on transcripts per million-normalized gene counts. Protein-coding genes expressed in ≥ 50% of samples with a q value < 0.05 and an absolute fold-change ≥ 2 were considered significantly differentially expressed. Bioinformatics in Ingenuity Pathway Analysis was performed to understand the biology of risk-associated expression profiles. Association was assessed between gene expression and risk via Pearson correlation analysis. Linear discriminant analysis models using significant genes were created and validated for classification of high-risk cases. RESULTS: We analyzed transcriptomes of 68 IA patients. In these cases, 31 IAs were large (≥ 5 mm), while 26 IAs had a high PHASES score. Based on size, 36 genes associated with high-risk IAs, and two were correlated with the size measurement. Alternatively, based on PHASES score, 76 genes associated with high-risk cases, and nine of them showed significant correlation to the score. Similar ontological terms were associated with both gene profiles, which reflected inflammatory signaling and vascular remodeling. Prediction models based on size and PHASES stratification were able to correctly predict IA risk status, with > 80% testing accuracy for both. CONCLUSIONS: Here, we identified genes associated with IA risk, as quantified by common clinical metrics. Preliminary classification models demonstrated feasibility of assessing IA risk using whole blood expression.

Hemodynamic Analysis Shows High Wall Shear Stress Is Associated with Intraoperatively Observed Thin Wall Regions of Intracranial Aneurysms.

BACKGROUND: Studying the relationship between hemodynamics and local intracranial aneurysm (IA) pathobiology can help us understand the natural history of IA. We characterized the relationship between the IA wall appearance, using intraoperative imaging, and the hemodynamics from CFD simulations. METHODS: Three-dimensional geometries of 15 IAs were constructed and used for CFD. Two-dimensional intraoperative images were subjected to wall classification using a machine learning approach, after which the wall type was mapped onto the 3D surface. IA wall regions included thick (white), normal (purple-crimson), and thin/translucent (red) regions. IA-wide and local statistical analyses were performed to assess the relationship between hemodynamics and wall type. RESULTS: Thin regions of the IA sac had significantly higher WSS, Normalized WSS, WSS Divergence and Transverse WSS, compared to both normal and thick regions. Thicker regions tended to co-locate with significantly higher RRT than thin regions. These trends were observed on a local scale as well. Regression analysis showed a significant positive correlation between WSS and thin regions and a significant negative correlation between WSSD and thick regions. CONCLUSION: Hemodynamic simulation results were associated with the intraoperatively observed IA wall type. We consistently found that elevated WSS and WSSNorm were associated with thin regions of the IA wall rather than thick and normal regions.

Aneurysm risk metrics and hemodynamics are associated with greater vessel wall enhancement in intracranial aneurysms.

Vessel wall enhancement (VWE) in contrast-enhanced magnetic resonance imaging (MRI) is a potential biomarker for intracranial aneurysm (IA) risk stratification. In this study, we investigated the relationship between VWE features, risk metrics, morphology and hemodynamics in 41 unruptured aneurysms. We reconstructed the IA geometries from MR angiography and mapped pituitary stalk-normalized MRI intensity on the aneurysm surface using an in-house tool. For each case, we calculated the maximum intensity (CRstalk) and IA risk (via size and the rupture resemblance score (RRS)). We performed correlation analysis to assess relationships between CRstalk and IA risk metrics (size and RRS), as well as each parameter encompassed in RRS, i.e. aneurysmal size ratio (SR), normalized wall shear stress (WSS) and oscillatory shear index. We found that CRstalk had a strong correlation (Pearson correlation coefficient, PCC = 0.630) with size and a moderate correlation (PCC = 0.472) with RRS, indicating an association between VWE and IA risk. Furthermore, CRstalk had a weak negative correlation with normalized WSS (PCC = -0.320) and a weak positive correlation with SR (PCC = 0.390). Local voxel-based analysis showed only a weak negative correlation between normalized WSS and contrast-enhanced MRI signal intensity (PCC = -0.240), suggesting that if low-normalized WSS induces enhancement-associated pathobiology, the effect is not localized.

An Image-Based Workflow for Objective Vessel Wall Enhancement Quantification in Intracranial Aneurysms.

BACKGROUND: VWE in contrast-enhanced magnetic resonance imaging (MRI) is a potential biomarker for the evaluation of IA. The common practice to identify IAs with VWE is mainly based on a visual inspection of MR images, which is subject to errors and inconsistencies. Here, we develop and validate a tool for the visualization, quantification and objective identification of regions with VWE. METHODS: N = 41 3D T1-MRI and 3D TOF-MRA IA images from 38 patients were obtained and co-registered. A contrast-enhanced MRI was normalized by the enhancement intensity of the pituitary stalk and signal intensities were mapped onto the surface of IA models generated from segmented MRA. N = 30 IAs were used to identify the optimal signal intensity value to distinguish the enhancing and non-enhancing regions (marked by an experienced neuroradiologist). The remaining IAs (n = 11) were used to validate the threshold. We tested if the enhancement area ratio (EAR-ratio of the enhancing area to the IA surface-area) could identify high risk aneurysms as identified by the ISUIA clinical score. RESULTS: A normalized intensity of 0.276 was the optimal threshold to delineate enhancing regions, with a validation accuracy of 81.7%. In comparing the overlap between the identified enhancement regions against those marked by the neuroradiologist, our method had a dice coefficient of 71.1%. An EAR of 23% was able to discriminate high-risk cases with an AUC of 0.7. CONCLUSIONS: We developed and validated a pipeline for the visualization and objective identification of VWE regions that could potentially help evaluation of IAs become more reliable and consistent.

Identification of intima-to-media signals for flow-induced vascular remodeling using correlative gene expression analysis.

Changes in blood flow can induce arterial remodeling. Intimal cells sense flow and send signals to the media to initiate remodeling. However, the nature of such intima-media signaling is not fully understood. To identify potential signals, New Zealand white rabbits underwent bilateral carotid ligation to increase flow in the basilar artery or sham surgery (n = 2 ligated, n = 2 sham). Flow was measured by transcranial Doppler ultrasonography, vessel geometry was determined by 3D angiography, and hemodynamics were quantified by computational fluid dynamics. 24 h post-surgery, the basilar artery and terminus were embedded for sectioning. Intima and media were separately microdissected from the sections, and whole transcriptomes were obtained by RNA-seq. Correlation analysis of expression across all possible intima-media gene pairs revealed potential remodeling signals. Carotid ligation increased flow in the basilar artery and terminus and caused differential expression of 194 intimal genes and 529 medial genes. 29,777 intima-media gene pairs exhibited correlated expression. 18 intimal genes had > 200 medial correlates and coded for extracellular products. Gene ontology of the medial correlates showed enrichment of organonitrogen metabolism, leukocyte activation/immune response, and secretion/exocytosis processes. This demonstrates correlative expression analysis of intimal and medial genes can reveal novel signals that may regulate flow-induced arterial remodeling.

Endogenous animal models of intracranial aneurysm development: a review.

The pathogenesis and natural history of intracranial aneurysm (IA) remains poorly understood. To this end, animal models with induced cerebral vessel lesions mimicking human aneurysms have provided the ability to greatly expand our understanding. In this review, we comprehensively searched the published literature to identify studies that endogenously induced IA formation in animals. Studies that constructed aneurysms (i.e., by surgically creating a sac) were excluded. From the eligible studies, we reported information including the animal species, method for aneurysm induction, aneurysm definitions, evaluation methods, aneurysm characteristics, formation rate, rupture rate, and time course. Between 1960 and 2019, 174 articles reported endogenous animal models of IA. The majority used flow modification, hypertension, and vessel wall weakening (i.e., elastase treatment) to induce IAs, primarily in rats and mice. Most studies utilized subjective or qualitative descriptions to define experimental aneurysms and histology to study them. In general, experimental IAs resembled the pathobiology of the human disease in terms of internal elastic lamina loss, medial layer degradation, and inflammatory cell infiltration. After the early 2000s, many endogenous animal models of IA began to incorporate state-of-the-art technology, such as gene expression profiling and 9.4-T magnetic resonance imaging (MRI) in vivo imaging, to quantitatively analyze the biological mechanisms of IA. Future studies aimed at longitudinally assessing IA pathobiology in models that incorporate aneurysm growth will likely have the largest impact on our understanding of the disease. We believe this will be aided by high-resolution, small animal, survival imaging, in situ live-cell imaging, and next-generation omics technology.

Landmark-Based Shape Analysis on Middle Cerebral Intracranial Aneurysms: A Geometric Morphometrics Approach to Infer Natural History.

BACKGROUND: Due to the scarcity of longitudinal data, the morphologic development of intracranial aneurysms (IAs) during their natural history remains poorly understood. However, longitudinal information can often be inferred from cross-sectional datasets as demonstrated by anatomists' use of geometric morphometrics to build evolutionary trees, reconstructing species inter-relationships based on morphologic landmarks. OBJECTIVE: We adopted these tools to analyze cross-sectional image data and infer relationships between IA morphologies. METHODS: On 3D reconstructions of 52 middle cerebral arteries (MCA) IAs (9 ruptured) and 10 IAfree MCAs (baseline geometries), 7 semi-automated landmarks were placed at the proximal parent artery and maximum height. From these, 64 additional landmarks were computationally generated to create a 71-landmark point cloud of 213 xyz coordinates. This data was normalized by Procrustes transformation and used in the principal component analysis, hierarchical clustering, and phylogenetic analyses. RESULTS: Principal component analysis showed separation of IA-free MCA geometries and grouping of ruptured IAs from unruptured IAs. Hierarchical clustering delineated a cluster of only unruptured IAs that were significantly smaller and more spherical than clusters that had ruptured IAs. Phylogenetic classification placed ruptured IAs more distally in the tree than unruptured IAs, indicating greater shape derivation. Groups of unruptured IAs were observed, but ruptured IAs were invariably found in mixed lineages with unruptured IAs, suggesting that some pathways of shape change may be benign while others are more associated with rupture. CONCLUSION: Geometric morphometric analyses of larger datasets may indicate particular pathways of shape change leading toward aneurysm rupture versus stabilization.

A data-driven model to identify high-risk aneurysms and guide management decisions: the Rupture Resemblance Score.

OBJECTIVE: Previous studies have found that ruptured intracranial aneurysms (RIAs) have distinct morphological and hemodynamic characteristics, including higher size ratio and oscillatory shear index and lower wall shear stress. Unruptured intracranial aneurysms (UIAs) that possess similar characteristics to RIAs may be at a higher risk of rupture than those UIAs that do not. The authors previously developed the Rupture Resemblance Score (RRS), a data-driven computer model that can objectively gauge the similarity of UIAs to RIAs in terms of morphology and hemodynamics. The authors aimed to explore the clinical utility of RRS in guiding the management of UIAs, especially for challenging cases such as small UIAs. METHODS: Between September 2018 and June 2019, the authors retrospectively collected consecutive challenging cases of incidentally identified UIAs that were discussed during their weekly multidisciplinary neurovascular conference. From patient 3D digital subtraction angiography, they reconstructed the aneurysm geometry and performed computer-assisted 3D morphology analysis and computational fluid dynamics simulation. They calculated RRS for every UIA case and compared it against the treatment decision made at the neurovascular conference as well as the recommendation based on the unruptured intracranial aneurysm treatment score (UIATS). RESULTS: Forty-seven patients with 79 UIAs, 90% of which were < 7 mm in size, were included in this study. The mean RRS (range 0.0-1.0) was 0.24 ± 0.31. At the conferences, treatment was endorsed for 45 of the UIAs (57%). These cases had significantly higher RRSs than the 34 cases suggested for observation (0.33 ± 0.34 vs 0.11 ± 0.19, p < 0.001). The UIATS-based recommendations were "observation" for 24 UIAs (30%), "treatment" for 21 UIAs (27%), and "not definitive" for 34 UIAs (43%). These "not definitive" cases were stratified by RRS based on similarity to RIAs. CONCLUSIONS: Although not a rupture predictor, RRS is a data-driven model that gauges the similarity of UIAs to RIAs in terms of morphology and hemodynamics. In cases in which the UIATS-based recommendation is not definitive, RRS provides additional stratification to assist the identification of high-risk UIAs. The current study highlights the clinical utility of RRS in a real-world setting as an adjunctive tool for the management of UIAs.

Inter-patient variations in flow boundary conditions at middle cerebral artery from 7T PC-MRI and influence on Computational Fluid Dynamics of intracranial aneurysms.

BACKGROUND: Computational fluid dynamics(CFD) of intracranial aneurysms requires flow boundary conditions(BCs) as inputs. Patient-specific BCs are usually unavailable and substituted by literature-derived generic BCs. Therefore, we investigated inter-patient BC variations and their influence on middle cerebral artery aneurysmal hemodynamics. METHOD: We retrospectively collected CT angiography and 7-T Phase-Contrast(PC)-MRI data from eight middle-cerebral-artery bifurcation aneurysms to reconstruct the geometry and measure the arterial flowrates, respectively. The coefficient of variation(CoV) was calculated for the inlet flowrate and the pulsatility index(PI). The outflow split estimated by Murray's law was compared with PC-MRI measurements. For each aneurysm, we performed seven simulations: "baseline" using PC-MRI-derived BCs and the other six with changing BCs to explore the influence of BC variations on hemodynamics. RESULTS: From PC-MRI, the inlet flowrate was 1.94 ± 0.71 cm3/s(CoV = 36%) and PI was 0.37 ± 0.13(CoV = 34%). The outflow split estimated by Murray's law deviated by 15.3% compared to PC-MRI. Comparing to "baseline" models, ±36% variations in inlet flowrate caused -61% to +89% changes in time-averaged wall shear stress(WSS), -37% to +32% in normalized WSS(NWSS; by parent-artery), and -42% to +126% in oscillatory shear index(OSI). The ±34% variations in PI caused, -46% to +67% in OSI. Applying ±15% variations in outflow split led to inflow jet deflection and -41% to +52% changes in WSS, -41% to +47% in NWSS, and -44% to +144% in OSI. CONCLUSION: Inflow rate and outflow split have a drastic impact on hemodynamics of intracranial aneurysms. Inlet waveform has a negligible impact on WSS and NWSS but major impact on OSI. CFD-based models need to consider such sensitivity.