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.

Objective To describe a novel case of coronavirus disease 2019 (COVID-19)-associated acute necrotizing encephalopathy (ANE) in a patient with aplastic anemia where there was early brain stem-predominant involvement. Methods Evaluation of cause, clinical symptoms, and treatment response. Results A 59-year-old woman with a background of transfusion-dependent aplastic anemia presented with seizures and reduced level of consciousness 10 days after the onset of subjective fever, cough, and headache. Nasopharyngeal swab testing for severe acute respiratory syndrome coronavirus (SARS-CoV-2) was positive, and CT during admission demonstrated diffuse swelling of the brain stem. She required intubation and mechanical ventilation for airway protection, given her reduced level of consciousness. The patient9s condition deteriorated, and MRI on day 6 demonstrated worsening brain stem swelling with symmetrical hemorrhagic lesions in the brain stem, amygdalae, putamina, and thalamic nuclei. Appearances were consistent with hemorrhagic ANE with early brain stem involvement. The patient showed no response to steroid therapy and died on the eighth day of admission. Conclusions COVID-19 may be associated with an acute severe encephalopathy and, in this case, was considered most likely to represent an immune-mediated phenomenon. As the pandemic continues, we anticipate that the spectrum of neurologic presentation will broaden. It will be important to delineate the full clinical range of emergent COVID-19-related neurologic disease.

Accurate and reliable intraoperative neuronavigation is crucial for achieving maximal safe resection of brain tumors. Intraoperative MRI (iMRI) has received significant attention as the next step in improving navigation. However, the immense cost and logistical challenge of iMRI precludes implementation in most centers worldwide. In comparison, intraoperative ultrasound (ioUS) is an affordable tool, easily incorporated into existing theatre infrastructure, and operative workflow. Historically, ultrasound has been perceived as difficult to learn and standardize, with poor, artifact-prone image quality. However, ioUS has dramatically evolved over the last decade, with vast improvements in image quality and well-integrated navigation tools. Advanced techniques, such as contrast-enhanced ultrasound (CEUS), have also matured and moved from the research field into actual clinical use. In this review, we provide a comprehensive and pragmatic guide to ioUS. A suggested protocol to facilitate learning ioUS and improve standardization is provided, and an outline of common artifacts and methods to minimize them given. The review also includes an update of advanced techniques and how they can be incorporated into clinical practice.

Background Cerebral microhaemorrhages are increasingly being recognised as a complication of COVID-19. This observational retrospective study aims to further investigate the potential pathophysiology through assessing the pattern of microhaemorrhage and clinical characteristics of patients with COVID-19 and microhaemorrhage. By comparing with similar patterns of microhaemorrhage in other non-COVID-19 disease, this study aims to propose possible common pathogenic mechanisms. Methods A retrospective observational case series was performed identifying all patients with COVID-19 complicated by cerebral microhaemorrhage on MRI. The distribution and number of microhaemorrhages were recorded using the microbleed anatomical scale, and patients’ baseline characteristics and salient test results were also recorded. Results Cerebral microhaemorrhages were noted to have a predilection for the corpus callosum, the juxtacortical white matter and brainstem. All patients had a preceding period of critical illness with respiratory failure and severe hypoxia necessitating intubation and mechanical ventilation. Discussion This study demonstrates a pattern of cerebral microhaemorrhage that is similar to the pattern reported in patients with non-COVID-19 related critical illness and other causes of severe hypoxia. This raises questions regarding whether microhaemorrhage occurs from endothelial dysfunction due the direct effect of SARS-CoV-2 infection or from the secondary effects of critical illness and hypoxia.

SUMMARY: There are very limited published data on the neurologic complications associated with coronavirus disease 2019 (COVID-19) in the pediatric population. Here we present the first 2 pediatric cases of presumed COVID-19 related cytotoxic lesions of the corpus callosum. Similar to reports in adults, these cases suggest that the COVID-19 infection in children may rarely mediate a hyperinflammatory response that can cause CNS pathology. As the pandemic continues further, the presentation of cytotoxic lesions of the corpus callosum should prompt radiologists to consider COVID-19, among other known causes.

Acute cerebrovascular disease, particularly ischaemic stroke, has emerged as a serious complication of COVID-19.1 However, the mechanism and optimal management of this remain incompletely understood. In the pulmonary and cardiac circulation, there is evidence that thrombotic complications may relate to endothelial inflammation and injury but evidence for this in the cerebrovascular system is limited.2 We demonstrate direct imaging evidence of vasculitis in a patient with COVID-19 complicated by multiple territory ischaemic strokes which responded to steroids and targeted interleukin-1 (IL-1) and interleukin-6 (IL-6) inhibition.