The assessment of rupture probability is crucial to identifying at risk intracranial aneurysms (IA) in patients harboring multiple aneurysms. We aimed to develop a computer-assisted detection system for small-sized aneurysm ruptures using a convolutional neural network (CNN) based on images of three-dimensional digital subtraction angiography. A retrospective data set, including 368 patients, was used as a training cohort for the CNN using the TensorFlow platform. Aneurysm images in six directions were obtained from each patient and the region-of-interest in each image was extracted. The resulting CNN was prospectively tested in 272 patients and the sensitivity, specificity, overall accuracy, and receiver operating characteristics (ROC) were compared to a human evaluator. Our system showed a sensitivity of 78.76% (95% CI: 72.30%–84.30%), a specificity of 72.15% (95% CI: 60.93%–81.65%), and an overall diagnostic accuracy of 76.84% (95% CI: 71.36%–81.72%) in aneurysm rupture predictions. The area under the ROC (AUROC) in the CNN was 0.755 (95% CI: 0.699%–0.805%), better than that obtained from a human evaluator (AUROC: 0.537; p < 0.001). The CNN-based prediction system was feasible to assess rupture risk in small-sized aneurysms with diagnostic accuracy superior to human evaluators. Additional studies based on a large data set are necessary to enhance diagnostic accuracy and to facilitate clinical application.

Park JC, et al. J Cerebrovasc Endovasc Neurosurg. 2017 Sep;19(3):184-188. https://doi.org/10.7461/jcen.2017.19.3.184

Auto-detection of cerebral aneurysms via convolutional neural network (CNN) is being increasingly reported. However, few studies to date have accurately predicted the risk, but not the diagnosis itself. We developed a multi-view CNN for the prediction of rupture risk involving small unruptured intracranial aneurysms (UIAs) based on three-dimensional (3D) digital subtraction angiography (DSA). The performance of a multi-view CNN-ResNet50 in accurately predicting the rupture risk (high vs. non-high) of UIAs in the anterior circulation measuring less than 7 mm in size was compared with various CNN architectures (AlexNet and VGG16), with similar type but different layers (ResNet101 and ResNet152), and single image-based CNN (single-view ResNet50). The sensitivity, specificity, and overall accuracy of risk prediction were estimated and compared according to CNN architecture. The study included 364 UIAs in training and 93 in test datasets. A multi-view CNN-ResNet50 exhibited a sensitivity of 81.82 (66.76–91.29)%, a specificity of 81.63 (67.50–90.76)%, and an overall accuracy of 81.72 (66.98–90.92)% for risk prediction. AlexNet, VGG16, ResNet101, ResNet152, and single-view CNN-ResNet50 showed similar specificity. However, the sensitivity and overall accuracy were decreased (AlexNet, 63.64% and 76.34%; VGG16, 68.18% and 74.19%; ResNet101, 68.18% and 73.12%; ResNet152, 54.55% and 72.04%; and single-view CNN-ResNet50, 50.00% and 64.52%) compared with multi-view CNN-ResNet50. Regarding F1 score, it was the highest in multi-view CNN-ResNet50 (80.90 (67.29–91.81)%). Our study suggests that multi-view CNN-ResNet50 may be feasible to assess the rupture risk in small-sized UIAs.

Mechanical thrombectomies with balloon-guide catheters (BGC) are thought to improve successful recanalization rates and to decrease the incidence of distal emboli compared to thrombectomies without BGC. We aimed to assess the effects of BGC on the outcomes of mechanical thrombectomy in acute ischemic strokes. Studies from PubMed, EMBASE, and the Cochrane library database from January 2010 to February 2018 were reviewed. Random effect model for meta-analysis was used. Analyses such as meta-regression and the "trim-and-fill" method were additionally carried out. A total of seven articles involving 2223 patients were analyzed. Mechanical thrombectomy with BGC was associated with higher rates of successful recanalization (odds ratio [OR], 1.632; 95% confidence interval [CI], 1.293-2.059). BGC did not significantly decrease distal emboli, both before (OR, 0.404; 95% CI, 0.108-1.505) and after correcting for bias (adjusted OR, 1.165; 95% CI, 0.310- 4.382). Good outcomes were observed more frequently in the BGC group (OR, 1.886; 95% CI, 1.564-2.273). Symptomatic intracranial hemorrhage and mortality did not differ significantly with BGC use. Our meta-analysis demonstrates that BGC enhance recanalization rates. However, BGC use did not decrease distal emboli after mechanical thrombectomies. This should be interpreted with caution due to possible publication bias and heterogeneity. Additional meta-analyses based on individual patient data are needed to clarify the role of BGC in mechanical thrombectomies.

Background: Although the dopaminergic system is interconnected with the hypothalamic-pituitary-thyroid axis, few studies have explained the causal relationship between thyroid disease and Parkinson’s disease (PD). Objective: The goal of this study was to investigate the association between thyroid diseases and PD in Korean residents. Methods: The Korean National Health Insurance Service-National Sample Cohort, which includes individuals aged ≥40 years, was assessed from 2002 to 2015. A total of 5,586 PD patients were matched by age, sex, income, and the region of residence with 22,344 control participants at a ratio of 1:4. In the PD and control groups, previous histories of levothyroxine treatment, goiter, hypothyroidism, thyroiditis, and hyperthyroidism were investigated. Results: The rates of levothyroxine treatment for more than 3 months, hypothyroidism, and hyperthyroidism were higher in the PD group than the control group (3.2%, 3.8%, and 2.8% vs. 2.5%, 2.9%, and 1.9%, respectively, p < 0.05). The adjusted odds ratios (ORs) in model 2, which was adjusted for all potential confounders, for hypothyroidism and hyperthyroidism in the PD group were 1.25 (95% confidence interval (CI) 1.01–1.55, p = 0.044) and 1.37 (95% CI 1.13–1.67, p = 0.002), respectively. In subgroup analyses, the association between hypothyroidism and PD was maintained in men older than 70 years and the association between hyperthyroidism and PD was maintained in women younger than 70 years. Conclusion: Both hyperthyroidism and hypothyroidism were associated with higher risk of PD, particularly for women younger than 70 years and men older than 70 years, respectively.

Objective To examine the safety and efficacy of mechanical thrombectomy using a retrievable stent for thromboembolic occlusion occurring during coil embolization of ruptured intracranial aneurysms. Methods Between June 2011 and June 2015, 631 consecutive patients with ruptured intracranial aneurysms underwent coil embolization at 6 hospitals. Among 53 patients who had thromboembolic complications, 15 patients harboring 15 aneurysms underwent rescue mechanical thrombectomy with a retrievable stent for the treatment of thromboembolic occlusion during the coiling of ruptured aneurysms. The patients' clinical and radiologic outcomes were retrospectively reviewed. Results Of the 15 aneurysms, coiling alone was used for 13 (86.7%), and stent-assisted coiling was performed for 2 (13.3%). Thromboembolic occlusion most frequently occurred distal to the aneurysm (n=10, 66.7%), followed by proximal to the aneurysm (n=3, 20%), and at the coil−parent vessel interface (n=2, 13.3%). All patients underwent mechanical thrombectomy with a retrievable stent, including 5 patients who were initially treated with an IA tirofiban infusion. Complete recanalization (Thrombolysis in Cerebral Infarction (TICI) 3) was obtained in 13 (86.7%) and partial recanalization (TICI 2b) in 2 (13.3%). Two patients who had received IA tirofiban before mechanical thrombectomy had hemorrhagic complications. At 6 months after discharge, 9 patients had a modified Rankin Scale (mRS) score of 1, 3 patients were mRS 2, 1 patient was mRS 3, 1 patient was mRS 4, and 1 patient was mRS 6. Conclusions Rescue mechanical thrombectomy using a retrievable stent can be a useful treatment for thromboembolic occlusion occurring during coil embolization of ruptured intracranial aneurysms.