BACKGROUND AND PURPOSE: Small intracranial aneurysms are being increasingly detected while the rupture risk is not well-understood. We aimed to develop rupture-risk models of small aneurysms by combining clinical, morphologic, and hemodynamic information based on machine learning techniques and to test the models in external validation datasets. MATERIALS AND METHODS: From January 2010 to December 2016, five hundred four consecutive patients with only small aneurysms (<5 mm) detected by CTA and invasive cerebral angiography (or surgery) were retrospectively enrolled and randomly split into training (81%) and internal validation (19%) sets to derive and validate the proposed machine learning models (support vector machine, random forest, logistic regression, and multilayer perceptron). Hemodynamic parameters were obtained using computational fluid dynamics simulation. External validation was performed in other hospitals to test the models. RESULTS: The support vector machine performed the best with areas under the curve of 0.88 (95% CI, 0.85–0.92) and 0.91 (95% CI, 0.74–0.98) in the training and internal validation datasets, respectively. Feature ranks suggested hemodynamic parameters, including stable flow pattern, concentrated inflow streams, and a small (<50%) flow-impingement zone, and the oscillatory shear index coefficient of variation, were the best predictors of aneurysm rupture. The support vector machine showed an area under the curve of 0.82 (95% CI, 0.69–0.94) in the external validation dataset, and no significant difference was found for the areas under the curve between internal and external validation datasets (P = .21). CONCLUSIONS: This study revealed that machine learning had a good performance in predicting the rupture status of small aneurysms in both internal and external datasets. Aneurysm hemodynamic parameters were regarded as the most important predictors.

Reconstruction of patient-specific biomechanical model of intracranial aneurysm has been based on different imaging modalities. However, different imaging techniques may influence the model geometry and the computational fluid dynamics (CFD) simulation. The aim of this study is to evaluate the differences of the morphological and hemodynamic parameters in the computational models reconstructed from computed tomography angiography (CTA), magnetic resonance angiography (MRA) and 3D rotational angiography (3DRA). Ten patients with cerebral aneurysms were enrolled in the study. MRA, CTA and 3DRA were performed on all patients. For each patient, three patient-specific models were reconstructed respectively based on the three sets of imaging data of the patient. CFD simulations were performed on each model. Model geometry and hemodynamic parameters were compared between the three models. In terms of morphological parameters, by comparing CTA based models (CM) and 3DRA based models (DM) which were treated as the “standard models”, the aspect ratio had the minimum difference (Δ = 8.3 ± 1.72 %, P = 0.953) and the surface distance was 0.25 ± 0.07 mm. Meanwhile, by comparing MRA based models (MM) and DM, the size had the minimum difference (Δ = 6.6 ± 1.85 %, P = 0.683) and the surface distance was 0.36 ± 0.1 mm. In respect of hemodynamic parameters, all three models showed a similar distribution: low average WSS at the sack, high OSI at the body and high average WSSG at the neck. However, there was a large variation in the average WSS (Δ = 34 ± 5.13 % for CM, Δ = 40.6 ± 9.21 % for MM). CTA and MRA have no significant differences in reproducing intracranial aneurysm geometry. The CFD results suggests there might be some significant differences in hemodynamic parameters between the three imaging-based models and this needs to be considered when interpreting the CFD results of different imaging-based models. If we only need to study the main flow patterns, three types of image-based model might be all suitable for patient-specific computational modeling studies.

BACKGROUND AND PURPOSE: Multidetector row CTA has become the primary imaging technique for detecting intracranial aneurysms. Technical progress enables the use of cerebral CTA with lower radiation doses and contrast media. We evaluated the diagnostic accuracy of 80-kV(peak) cerebral CTA with 30 mL of contrast agent for detecting intracranial aneurysms. MATERIALS AND METHODS: Two hundred four patients were randomly divided into 2 groups. Patients in group A (n = 102) underwent 80-kVp CTA with 30 mL of contrast agent, while patients in group B (n = 102) underwent conventional CTA (120 kVp, 60 mL of contrast agent). All patients underwent DSA. Image quality, diagnostic accuracy, and radiation dose between the 2 groups were compared. RESULTS: Diagnostic image quality was obtained in 100 and 99 patients in groups A and B, respectively (P = .65). With DSA as reference standard, diagnostic accuracy on a per-aneurysm basis was 89.9% for group A and 93.9% for group B. For evaluating smaller aneurysms ( .05) or between the 2 CTA groups (all, P > .05). The effective dose in group A was reduced by 72.7% compared with group B. CONCLUSIONS: In detecting intracranial aneurysms with substantial radiation dose and contrast agent reduction, 80-kVp/30-mL contrast CTA provides the same diagnostic accuracy as conventional CTA.

Hemodynamics derived from computational fluid dynamics based on magnetic resonance angiography is associated with functional outcomes in atherosclerotic middle cerebral artery stenosis Jiahua Wu, Peng Wang, Leilei Zhou, Danfeng Zhang, Qian Chen, Cunnan Mao, Wen Su, Yingsong Huo, Jin Peng, Xindao Yin, Guozhong Chen