Although machine learning (ML) has shown promise across disciplines, out-of-sample generalizability is concerning. This is currently addressed by sharing multi-site data, but such centralization is challenging/infeasible to scale due to various limitations. Federated ML (FL) provides an alternative paradigm for accurate and generalizable ML, by only sharing numerical model updates. Here we present the largest FL study to-date, involving data from 71 sites across 6 continents, to generate an automatic tumor boundary detector for the rare disease of glioblastoma, reporting the largest such dataset in the literature (n = 6, 314). We demonstrate a 33% delineation improvement for the surgically targetable tumor, and 23% for the complete tumor extent, over a publicly trained model. We anticipate our study to: 1) enable more healthcare studies informed by large diverse data, ensuring meaningful results for rare diseases and underrepresented populations, 2) facilitate further analyses for glioblastoma by releasing our consensus model, and 3) demonstrate the FL effectiveness at such scale and task-complexity as a paradigm shift for multi-site collaborations, alleviating the need for data-sharing.

Computational competitions are the standard for benchmarking medical image analysis algorithms, but they typically use small curated test datasets acquired at a few centers, leaving a gap to the reality of diverse multicentric patient data. To this end, the Federated Tumor Segmentation (FeTS) Challenge represents the paradigm for real-world algorithmic performance evaluation. The FeTS challenge is a competition to benchmark (i) federated learning aggregation algorithms and (ii) state-of-the-art segmentation algorithms, across multiple international sites. Weight aggregation and client selection techniques were compared using a multicentric brain tumor dataset in realistic federated learning simulations, yielding benefits for adaptive weight aggregation, and efficiency gains through client sampling. Quantitative performance evaluation of state-of-the-art segmentation algorithms on data distributed internationally across 32 institutions yielded good generalization on average, albeit the worst-case performance revealed data-specific modes of failure. Similar multi-site setups can help validate the real-world utility of healthcare AI algorithms in the future.

Moyamoya disease is a chronic occlusive cerebrovascular disease that is non-inflammatory and non-atherosclerotic. It is characterized by endothelial hyperplasia and fibrosis of the intracranial portion of the carotid artery and its proximal branches, leading to progressive stenosis and occlusion, often clinically manifesting as ischemic or hemorrhagic stroke with high rates of morbidity and mortality. On cerebral angiography, the formation of collateral vessels has the appearance of a puff of smoke (moyamoya in Japanese), which became more conspicuous with the refinement of modern imaging techniques. When there is associated disease, it is known as moyamoya syndrome. Treatments are currently limited, although surgical revascularization may prevent ischemic events and preserve quality of life. In this review, we summarize recent advances in moyamoya disease, covering aspects of epidemiology, etiology, presentation, imaging, and treatment strategies.

SUMMARY: Glioneuronal tumors are characterized exclusively by neurocytic elements (neuronal tumors) or a combination of neuronal and glial features (mixed neuronal-glial tumors). Most of these tumors occur in young patients and are related to epilepsy. While ganglioglioma, dysembryoplastic neuroepithelial tumor, and desmoplastic infantile tumor are common glioneuronal tumors, anaplastic ganglioglioma, papillary glioneuronal tumor, rosette-forming glioneuronal tumor, gangliocytoma, and central neurocytoma are less frequent. Advances in immunohistochemical and molecular diagnostics have improved the characterization of these tumors and favored the description of variants and new subtypes, some not yet classified by the World Health Organization. Not infrequently, the histologic findings of biopsies of glioneuronal tumors simulate low-grade glial neoplasms; however, some imaging findings favor the correct diagnosis, making neuroimaging essential for proper management. Therefore, the aim of this review was to present key imaging, histopathology, immunohistochemistry, and molecular findings of glioneuronal tumors and their variants.

Background: The rapid evaluation of non-contrast-enhanced computed tomography (NCCT) brain scans in patients with anterior stroke symptoms saves time and favors optimal and prompt treatment. e-ASPECTS is a tool that automatically calculates the Alberta Stroke Program Early CT Score (ASPECTS) values, leading to a more accurate and timely image evaluation. Objective: To determine the ability of e-ASPECTS in differentiating images with and without injury. Methods: One-hundred sixteen patients admitted to a stroke unit in a Brazilian tertiary hospital underwent a CT scan at admission and at least one control brain imaging (NCCT or magnetic resonance imaging - MRI) 24 hours after admission. ASPECTS evaluation was performed by three neuroradiologists, three neurologists, and three neurology residents, all blinded to the symptoms and the injury side. The scores were compared to the ground truth, and an ASPECTS score was provided by two independent non blinded evaluators. Sensitivity and specificity were analyzed, and receiver operating characteristic curves, Bland-Altman plots with mean error score, and Matthews correlation coefficients (MCCs) were obtained for ASPECTS scores, assuming values equal to 10 for images without injury and values other than 10 for images with ischemic injury. Results: e-ASPECTS demonstrated similar performance to that of neuroradiologists and neurologists, with an area under the curve of 0.78 and an MCC value of 0.48 in the dichotomous analysis. The sensitivity and specificity of e-ASPECTS were 75% and 73%, respectively. Conclusion: e-ASPECTS is a validated and reliable tool for determining early signs of ischemia in NCCT.

Tumor factors that influence vestibular function in vestibular schwannoma (VS) have not been properly described. We evaluated whether cystic VSs have different vestibular function than solid VS. Tumor size on vestibular function was also evaluated. Cross-sectional study. Tertiary referral center. Forty-one cases of sporadic, untreated VS. Evaluation with video head impulse test and MRI. Tumors were classified as solid, heterogeneous, or cystic and by size using the Hannover classification. Vestibulo-ocular reflex (VOR) gain was correlated to tumor size and cystic status. Large VS had worse VOR gain than small lesions (p < 0.001). Cystic lesions had lower VOR gain than all other tumors (p = 0.001), Hannover T3 and T4 (p = 0.014), Hannover T4 (p = 0.015), solid tumors (p < 0.001), solid Hannover T3 and T4 (p = 0.003), and solid Hannover T4 (p = 0.008). Heterogeneous VSs had lower VOR gain compared to solid tumors (p = 0.02), solid Hannover T3 and T4 (p = 0.08), and solid Hannover T4 (p = 0.14). Heterogeneous and cystic VSs had lower VOR gain than solid tumors (p < 0.001), solid Hannover T3 and T4 (p = 0.004), and solid Hannover T4 (p = 0.02). VOR gain of solid T4 lesions was not significantly lower than solid Hannover T1-T3 (p = 0.33). Cystic status is directly associated with a worse vestibular dysfunction. Size did not significantly impact vestibular function in solid VS.