Posterior fossa tumors (PFT) are the most common tumors in children. Differentiation between the various PFT types is critical, as different tumors have diverse treatment approaches. This study proposes the use of fused architecture comprising two neural networks, a pre-trained ResNet-50 Convolutional Neural Network (CNN) and a tabular based network for the classification of PFT. The study included data for 158 MRI scans of 22 healthy controls and 136 pediatric patients with newly diagnosed PFT (63 Pilocytic Astrocytoma, 57 Medulloblastoma and 16 Ependymoma). The input data for classification were from magnetic resonance imaging: post contrast T1-weighted, fluid attenuated inversion recovery and diffusion Trace images, and tabular data: subject's age. Evaluation of the model was performed in a stratified 5-fold cross-validation manner, based on accuracy, precision, recall and F1 score metrics. Model explanation was performed in terms of visual explanation of the CNN by Gradient-weighted Class Activation Mapping (Grad-CAM) and by testing the contribution to the classification results of the different imaging input data sets and the proposed fused architectures relative to CNN only and tabular only architectures. The best classification results were obtained with the fused CNN + tabular data architecture, and based on diffusion Trace images, achieving mean cross-validation accuracy of 0.88 ± 0.04 for the validation and 0.87 ± 0.02 for the test dataset. Overall, the proposed architecture achieved improvement in accuracy and F1 score compared to CNN method for this dataset. The source code is available on the GitHub repository: https://github.com/artzimy/CNNTabular.
BACKGROUND: Fetal-type posterior cerebral arteries (F-PCAs) might result in alterations in hemodynamic flow patterns and may predispose an individual to an increased risk of posterior communicating artery aneurysms (PCoAAs). OBJECTIVE: To determine the association between PCoAAs and the presence of ipsilateral F-PCAs. METHODS: We retrospectively reviewed the radiographic findings from 185 patients harboring 199 PCoAAs that were treated at our institution between 2005 and 2015. Our study population consisted of 4 cohorts: (A) patients with 171 internal carotid arteries (ICAs) harboring unilateral PCoAAs; (B) 171 unaffected ICAs in the same patients from the first group; (C) 28 ICAs of 14 patients with bilateral PCoAAs; and (D) 180 ICAs of 90 patients with aneurysms in other locations. We then determined the presence of ipsilateral F-PCAs and recorded all aneurysm characteristics. RESULTS: Group A had the highest prevalence of F-PCAs (42%) compared to 19% in group B, 3% in group C, and 14% in group D (odds ratio A : B = 3.041; A : C = 19.626; and A : D = 4.308; P < .001). PCoAAs were associated with larger diameters of the posterior communicating arteries (median value 1.05 vs 0.86 mm; P = .001). The presence of F-PCAs was associated with larger sizes of the aneurysm necks (median value 3.3 vs 3.0 mm; P = .02). CONCLUSION: PCoAAs were associated with a higher prevalence of ipsilateral F-PCAs. This variant was associated with larger sizes of the aneurysm necks but was not associated with the sizes of the aneurysm domes or with their rupture statuses.
OBJECTIVE: The purpose of this study was to analyze the impact of adenosine-induced cardiac arrest (AiCA) on temporary clipping (TC) and the postoperative cerebral infarction rate among patients undergoing intracranial aneurysm surgery.METHODS: In this retrospective matched-cohort study, 65 patients who received adenosine for decompression of aneurysms during microsurgical clipping were identified (Group A) and randomly matched with 65 selected patients who underwent clipping but did not receive adenosine during surgery (Group B). The matching criteria included age, Fisher grade, aneurysm size, rupture status, and location of aneurysms. The primary outcomes were TC time and the postoperative infarction rate. The secondary outcome was the incidence of intraoperative aneurysm rupture (IAR).RESULTS: In Group A, 40 patients underwent clipping with AiCA alone and 25 patients (38%) received AiCA combined with TC, and in Group B, 60 patients (92%) underwent aneurysm clipping under the protection of TC (OR 0.052; 95% CI 0.018–0.147; p < 0.001). Group A required less TC time (2.04 minutes vs 4.46 minutes; p < 0.001). The incidence of postoperative lacunar infarction was equal in both groups (6.2%). There was an insignificant between-group difference in the incidence of IAR (1.5% in Group A vs 6.1% in Group B; OR 0.238; 95% CI 0.026–2.192; p = 0.171).CONCLUSIONS: AiCA is a useful technique for microneurosurgical treatment of cerebral aneurysms. AiCA can minimize the use of TC and does not increase the risk of IAR and postoperative infarction.