Accuracy and reliability of computer-assisted semi-automated morphological analysis of intracranial aneurysms: an experimental study with digital phantoms and clinical aneurysm cases

Combining Machine-Measured Morphometric, Geometric, and Hemodynamic Factors to Predict the Risk of Aneurysm Rupture at the Middle Cerebral Artery Bifurcation

BACKGROUND

Middle Cerebral Artery Bifurcation Aneurysm (MbifA) is associated with a high risk of rupture and poor overall prognosis in patients once it ruptures. Morphological, geometric, and hemodynamic parameters have been identified as factors contributing to the rupture of intracranial aneurysms. However, there are no studies that combine these 3 types of parameters to specifically target MbifA rupture.

METHODS

This study enrolled all patients with MbifAs diagnosed at our treatment center from 1 April 2021 to 31 July 2023 who met the study criteria. All patients underwent digital subtraction angiography examination to obtain 3D rotational angiography data. We imported the complete image data into the Aneurysm/Artery Reconstruction and Analysis machine to obtain 13 morphological parameters (Dneck, Ddome, Height, Dmax, Dartery, aspect ratio [AR], size ratio, dome-neck-ratio [DNR], height-artery-ratio, bottleneck factor, Inflow Angle, Incline Angle, Arterial Angle), 5 geometric parameters (V,S,undulation index [UI], ellipticity index [EI],nonsphericity index [NSI]), and 5 hemodynamic parameters (wall shear stress [WSS], the maximum WSS, the parent artery WSS, the normalized WSS [NWSS], oscillatory shear index [OSI]). All the above significant parameters were tested by univariate and multivariate analyses to find out the independent discriminatory factors.

RESULTS

A total of 49 MbifAs (16 ruptured and 33 unruptured) from 44 patients were included in the study. Height (P = 0.033), AR (P = 0.007), DNR (P = 0.011), EI (P = 0.042), NSI(P = 0.030), UI(P = 0.027), WSS(P = 0.033), and NWSS(P = 0.002) were all associated with MbifA rupture in univariate analyses, but only NWSS was an independent risk factor (P = 0.036, OR = 0.046, 95% CI: 0.003-0.815) in multivariate logistic regression analysis.

CONCLUSIONS

Height, AR, DNR, EI, UI, NSI, WSS, and NWSS may be correlated with MbifA rupture, but only NWSS was an independent risk factor. A lower NWSS was associated with a higher risk of MbifA rupture. No significant differences were observed in the angle parameters, including the Inflow Angle, between ruptured and unruptured MbifAs. OSI was significantly increased at the dome of the aneurysm but the mean OSI was not found to be associated with MbifA rupture.

Risk factors of unruptured intracranial aneurysms instability in the elderly

BACKGROUND

Presently, a consistent strategy for determining the stability of unruptured intracranial aneurysms (UIAs) in elderly patients is lacking, primarily due to the unique characteristics of this demographic. Our objective was to assess the risk factors contributing to aneurysm instability (growth or rupture) within the elderly population.

METHODS

In this study, we compiled data from follow-up patients with UIAs spanning from November 2016 to August 2021. We specifically focused on patients aged ≥ 60 years. Clinical histories were gathered, and morphological parameters of aneurysms were measured. The growth of aneurysms was determined using the computer-assisted semi-automated measurement (CASAM). Growth and rupture rates of UIAs were calculated, and both univariate and multivariate Cox regression analyses were conducted. Additionally, Kaplan-Meier survival curves were plotted.

RESULTS

A total of 184 patients with 210 aneurysms were enrolled in the study. The follow-up period encompasses 506.6 aneurysm-years and 401.4 patient-years. Among all the aneurysms, 23 aneurysms exhibited growth, with an annual aneurysm growth rate of 11.0%, and 1 (4.5%) experienced rupture, resulting in an annual aneurysm rupture rate of 0.21%. Multivariate Cox analysis identified poorly controlled hypertension (P = 0.011) and high-risk aneurysms (including anterior cerebral artery (ACA), anterior communicating artery (AcoA), posterior communicating artery aneurysm (PcoA), posterior circulation (PC) > 4 mm or distal internal carotid artery (ICAd), middle cerebral artery (MCA), and PC > 7 mm) (P = 0.006) as independent risk factors for the development of unstable aneurysms.

CONCLUSIONS

In the elderly, poorly controlled hypertension and high-risk aneurysms emerge as significant risk factors for aneurysm instability. This underscores the importance of rigorous surveillance or timely intervention in patients presenting with these risk factors.

Computer-assisted microcatheter shaping for intracranial aneurysm embolization: evaluation of safety and efficacy in a multicenter randomized controlled trial

BACKGROUND

This study aimed to evaluate the efficacy, stability, and safety of computer-assisted microcatheter shaping (CAMS) in patients with intracranial aneurysms.

METHODS

A total of 201 patients with intracranial aneurysms receiving endovascular coiling therapy were continuously recruited and randomly assigned to the CAMS and manual microcatheter shaping (MMS) groups. The investigated outcomes included the first-trial success rate, time to position the microcatheter in aneurysms, rate of successful microcatheter placement within 5 min, delivery times, microcatheter stability, and delivery performance.

RESULTS

The rates of first-trial success (96.0% vs 66.0%, P<0.001), successful microcatheter placement within 5 min (96.04% vs 72.00%, P<0.001), microcatheter stability (97.03% vs 84.00%, P=0.002), and 'excellent' delivery performance (45.54% vs 24.00%, P<0.001) in the CAMS group were significantly higher than those in the MMS group. Additionally, the total microcatheter delivery and positioning time (1.05 minutes (0.26) vs 1.53 minutes (1.00)) was significantly shorter in the CAMS group than in the MMS group (P<0.001). Computer assistance (OR 14.464; 95% CI 4.733 to 44.207; P<0.001) and inflow angle (OR 1.014; 95% CI 1.002 to 1.025; P=0.021) were independent predictors of the first-trial success rate. CAMS could decrease the time of microcatheter position compared with MMS, whether for junior or senior surgeons (P<0.001). Moreover, computer assistance technology may be more helpful in treating aneurysms with acute angles (p<0.001).

CONCLUSIONS

The use of computer-assisted procedures can enhance the efficacy, stability, and safety of surgical plans for coiling intracranial aneurysms.

Clinical, 3D Morphological, and Hemodynamic Risk Factors for Instability of Unruptured Intracranial Aneurysms

PURPOSE

Neurosurgeons can manage unruptured intracranial aneurysms (UIAs). However, the stability of UIAs under follow-up remains uncertain. This study aimed to examine the risk factors associated with the instability (rupture or growth) of UIAs during follow-up.

METHODS

We obtained information on patients with UIA who underwent ≥ 6 months of the time of flight-magnetic resonance angiography (TOF-MRA) imaging follow-up in two centers. Computer-assisted semi-automated measurement (CASAM) techniques were used for recording morphological parameters and determining the growth of these aneurysms. We also recorded hemodynamic parameters at the beginning of the follow-up. The univariate and multivariate Cox regression analyses were performed to calculate hazard ratios with corresponding 95% confidence intervals for the clinical, morphological, and hemodynamic risk factors for aneurysm instability.

RESULTS

A total of 304 aneurysms from 263 patients (80.4%) were included for analysis. The annual aneurysm growth rate was 4.7%. Significant predictive factors for aneurysm instability in the multivariate analysis were as follows: poorly controlled hypertension (hazard ratio (HR), 2.97 (95% CI, 1.27-6.98), P = 0.012); aneurysms located on posterior circulation (HR, 7.81 (95% CI, 2.28-26.73), P = 0.001), posterior communication artery (HR, 3.01 (95% CI, 1.07-8.46), P = 0.036), and cavernous carotid artery (HR, 3.78 (95% CI, 1.18-12.17), P = 0.026); and size ratio ≥ 0.87 (HR, 2.54 (95% CI, 1.14-5.68), P = 0.023).

CONCLUSIONS

The management of UIAs should focus on the control of hypertension during the follow-up. Aneurysms on the posterior communicating artery, posterior circulation, and cavernous carotid arteries require intensive surveillance or timely treatment.

Development and validation of a deep learning model for prediction of intracranial aneurysm rupture risk based on multi-omics factor

OBJECTIVE

The clinical ability of radiomics to predict intracranial aneurysm rupture risk remains unexplored. This study aims to investigate the potential uses of radiomics and explore whether deep learning (DL) algorithms outperform traditional statistical methods in predicting aneurysm rupture risk.

METHODS

This retrospective study included 1740 patients with 1809 intracranial aneurysms confirmed by digital subtraction angiography at two hospitals in China from January 2014 to December 2018. We randomly divided the dataset (hospital 1) into training (80%) and internal validation (20%). External validation was performed using independent data collected from hospital 2. The prediction models were developed based on clinical, aneurysm morphological, and radiomics parameters by logistic regression (LR). Additionally, the DL model for predicting aneurysm rupture risk using integration parameters was developed and compared with other models.

RESULTS

The AUCs of LR models A (clinical), B (morphological), and C (radiomics) were 0.678, 0.708, and 0.738, respectively (all p < 0.05). The AUCs of the combined feature models D (clinical and morphological), E (clinical and radiomics), and F (clinical, morphological, and radiomics) were 0.771, 0.839, and 0.849, respectively. The DL model (AUC = 0.929) outperformed the machine learning (ML) (AUC = 0.878) and the LR models (AUC = 0.849). Also, the DL model has shown good performance in the external validation datasets (AUC: 0.876 vs 0.842 vs 0.823, respectively).

CONCLUSION

Radiomics signatures play an important role in predicting aneurysm rupture risk. DL methods outperformed conventional statistical methods in prediction models for the rupture risk of unruptured intracranial aneurysms, integrating clinical, aneurysm morphological, and radiomics parameters.

KEY POINTS

• Radiomics parameters are associated with the rupture risk of intracranial aneurysms. • The prediction model based on integrating parameters in the deep learning model was significantly better than a conventional model. • The radiomics signature proposed in this study could guide clinicians in selecting appropriate patients for preventive treatment.

Artificial Intelligence Applications in Intracranial Aneurysm: Achievements, Challenges and Opportunities

Intracranial aneurysms present in about 3% of the general population and the number of detected aneurysms is continuously rising with the advances in imaging techniques. Intracranial aneurysm rupture carries a high risk of death or permanent disabilities; therefore assessment of the intracranial aneurysm along the entire course is of great clinical importance. Given the outstanding performance of artificial intelligence (AI) in image-based tasks, many AI-based applications have emerged in recent years for the assessment of intracranial aneurysms. In this review we will summarize the state-of-the-art of AI applications in intracranial aneurysms, emphasizing the achievements, and exploring the challenges. We will also discuss the future prospects and potential opportunities. This article provides an updated view of the AI applications in intracranial aneurysms and may act as a basis for guiding the related future works.

Case Report: Dynamic Changes in Hemodynamics During the Formation and Progression of Intracranial Aneurysms

Despite the devastating consequences of aneurysmal subarachnoid hemorrhage (SAH), the mechanisms underlying the formation, progression, and rupture of intracranial aneurysms (IAs) are complex and not yet fully clear. In a real-world situation, continuously observing the process of aneurysm development in humans appears unrealistic, which also present challenges for the understanding of the underlying mechanism. We reported the relatively complete course of IA development in two real patients. On this basis, computational fluid dynamics simulation (CFD) was performed to evaluate the changes in hemodynamics and analyze the mechanism underlying the formation, progression, and rupture of IAs. Our results suggested that the formation and progression of IAs can be a dynamic process, with constantly changing hemodynamic characteristics. CFD analysis based on medical imaging provides the opportunity to study the hemodynamic conditions over time. From these two rare cases, we found that concentrated high-velocity inflow jets, flows with vortex structures, extremely high WSS, and a very steep WSSG were correlated with the formation of IAs. Complex multi-vortex flows are possibly related to IAs prior to growth, and the rupture of IAs is possibly related to low WSS, extreme instability and complexity of flow patterns. Our findings provide unique insight into the theoretical hemodynamic mechanism underlying the formation and progression of IAs. Given the small sample size the findings of this study have to be considered preliminary and exploratory.

Risk Factors for Pericallosal Artery Aneurysm Rupture Based on Morphological Computer-Assisted Semiautomated Measurement and Hemodynamic Analysis

Background: Although pericallosal artery aneurysms (PAAs) are relatively uncommon, accounting for only 1-9% of all intracranial aneurysms (IAs), they exhibit a considerably high propensity to rupture. Nevertheless, our current knowledge of the risk factors for PAA rupture is still very limited. To fill this gap, we investigated rupture risk factors for PAAs based on morphological computer-assisted semiautomated measurement (CASAM) and hemodynamic analysis. Methods: Patients with PAAs were selected from the IA database in our institute and their baseline data were collected. Morphological parameters were measured in all enrolled patients by applying CASAM. Computational fluid dynamics simulation (CFD) was performed to evaluate the hemodynamic difference between ruptured and unruptured PAAs. Results: From June 2017 to June 2020, among 2141 patients with IAs in our institute, 47 had PAAs (2.2%). Thirty-one patients (mean age 57.65 ± 9.97 years) with 32 PAAs (20 unruptured and 12 ruptured) were included in the final analysis. Comparing with unruptured PAAs, ruptured PAAs had significantly higher aspect ratio (AR), mean normalized wall shear stress (NWSS), and mean oscillatory shear index (OSI) values than the unruptured PAAs (all P < 0.05) in univariate analyses. Multivariable analysis showed that a high mean OSI was an independent risk factor for PAA rupture (OR = 6.45, 95% CI 1.37-30.32, P = 0.018). Conclusion: This preliminary study indicates that there are morphological and hemodynamic differences between ruptured and unruptured PAAs. In particular, a high mean OSI is an independent risk factor for PAA rupture. Further research with a larger sample size is warranted in the future.