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.

Background Subarachnoid hemorrhage from cerebral aneurysm rupture is a major cause of morbidity and mortality. Early aneurysm identification, aided by automated systems, may improve patient outcomes. Therefore, a systematic review and meta-analysis of the diagnostic accuracy of artificial intelligence (AI) algorithms in detecting cerebral aneurysms using CT, MRI or DSA was performed. Methods MEDLINE, Embase, Cochrane Library and Web of Science were searched until August 2021. Eligibility criteria included studies using fully automated algorithms to detect cerebral aneurysms using MRI, CT or DSA. Following Preferred Reporting Items for Systematic Reviews and Meta-Analysis: Diagnostic Test Accuracy (PRISMA-DTA), articles were assessed using Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Meta-analysis included a bivariate random-effect model to determine pooled sensitivity, specificity, and area under the receiver operator characteristic curve (ROC-AUC). PROSPERO: CRD42021278454. Results 43 studies were included, and 41/43 (95%) were retrospective. 34/43 (79%) used AI as a standalone tool, while 9/43 (21%) used AI assisting a reader. 23/43 (53%) used deep learning. Most studies had high bias risk and applicability concerns, limiting conclusions. Six studies in the standalone AI meta-analysis gave (pooled) 91.2% (95% CI 82.2% to 95.8%) sensitivity; 16.5% (95% CI 9.4% to 27.1%) false-positive rate (1-specificity); 0.936 ROC-AUC. Five reader-assistive AI studies gave (pooled) 90.3% (95% CI 88.0% – 92.2%) sensitivity; 7.9% (95% CI 3.5% to 16.8%) false-positive rate; 0.910 ROC-AUC. Conclusion AI has the potential to support clinicians in detecting cerebral aneurysms. Interpretation is limited due to high risk of bias and poor generalizability. Multicenter, prospective studies are required to assess AI in clinical practice.

SUMMARY: As we defeat infectious diseases and cancer, one of the greatest medical challenges facing us in the mid-21st century will be the increasing prevalence of degenerative disease. Those diseases, which affect movement and cognition, can be the most debilitating. Dysfunction of the extrapyramidal system results in increasing motor disability often manifest as tremor, bradykinesia, and rigidity. The common pathologic pathway of these diseases, collectively described as parkinsonian syndromes, such as Parkinson disease, multiple system atrophy, progressive supranuclear palsy, corticobasal degeneration, and dementia with Lewy bodies, is degeneration of the presynaptic dopaminergic pathways in the basal ganglia. Conventional MR imaging is insensitive, especially in early disease, so functional imaging has become the primary method used to differentiate a true parkinsonian syndrome from vascular parkinsonism, drug-induced changes, or essential tremor. Unusually for a modern functional imaging technique, the method most widely used in European clinics depends on SPECT and not PET. This SPECT technique (described in the first of 2 parts) commonly reports dopamine-transporter function, with decreasing striatal uptake demonstrating increasingly severe disease.

Objective: Summarize evidence for use of advanced MRI techniques as monitoring biomarkers in the clinic, and highlight the latest bench-to-bedside developments.Methods: Experts in advanced MRI techniques applied to high-grade glioma treatment response assessment convened through a European framework. Current evidence regarding the potential for monitoring biomarkers in adult high-grade glioma is reviewed, and individual modalities of perfusion, permeability, and microstructure imaging are discussed (in Part 1 of two). In Part 2, we discuss modalities related to metabolism and/or chemical composition, appraise the clinic readiness of the individual modalities, and consider post-processing methodologies involving the combination of MRI approaches (multiparametric imaging) or machine learning (radiomics).Results: High-grade glioma vasculature exhibits increased perfusion, blood volume, and permeability compared with normal brain tissue. Measures of cerebral blood volume derived from dynamic susceptibility contrast-enhanced MRI have consistently provided information about brain tumor growth and response to treatment; it is the most clinically validated advanced technique. Clinical studies have proven the potential of dynamic contrast-enhanced MRI for distinguishing post-treatment related effects from recurrence, but the optimal acquisition protocol, mode of analysis, parameter of highest diagnostic value, and optimal cut-off points remain to be established. Arterial spin labeling techniques do not require the injection of a contrast agent, and repeated measurements of cerebral blood flow can be performed. The absence of potential gadolinium deposition effects allows widespread use in pediatric patients and those with impaired renal function. More data are necessary to establish clinical validity as monitoring biomarkers. Diffusion-weighted imaging, apparent diffusion coefficient analysis, diffusion tensor or kurtosis imaging, intravoxel incoherent motion, and other microstructural modeling approaches also allow treatment response assessment; more robust data are required to validate these alone or when applied to post-processing methodologies.Conclusion: Considerable progress has been made in the development of these monitoring biomarkers. Many techniques are in their infancy, whereas others have generated a larger body of evidence for clinical application.

A liver abscess is a feared and potentially fatal complication following transarterial chemoembolisation (TACE) and radiofrequency ablation (RFA) of liver tumours. Iatrogenic bilio-enteric communications, such as bilio-enteric anastomosis, sphincterotomy and biliary stents, are considered major risk factors and are due to bacterial colonisation of the biliary tree with enteric flora. Naturally occurring spontaneous cholecysto-enteric fistula poses a similar risk as its iatrogenic counterparts but is rarely described in the literature. We present a case where abscess formation complicated a combined TACE and RFA in an unrecognised cholecystocolic fistula.

Background: The impact on clinical outcomes of patient selection using perfusion imaging for endovascular thrombectomy (EVT) in patients with acute ischemic stroke presenting beyond 6 hours from onset remains undetermined in routine clinical practice. Methods: Patients from a national stroke registry that underwent EVT selected with or without perfusion imaging (noncontrast computed tomography/computed tomography angiography) in the early (<6 hours) and late (6–24 hours) time windows, between October 2015 and March 2020, were compared. The primary outcome was the ordinal shift in the modified Rankin Scale score at hospital discharge. Other outcomes included functional independence (modified Rankin Scale score ≤2) and in-hospital mortality, symptomatic intracerebral hemorrhage, successful reperfusion (Thrombolysis in Cerebral Infarction score 2b–3), early neurological deterioration, futile recanalization (modified Rankin Scale score 4–6 despite successful reperfusion) and procedural time metrics. Multivariable analyses were performed, adjusted for age, sex, baseline stroke severity, prestroke disability, intravenous thrombolysis, mode of anesthesia (Model 1) and including EVT technique, balloon guide catheter, and center (Model 2). Results: We included 4249 patients, 3203 in the early window (593 with perfusion versus 2610 without perfusion) and 1046 in the late window (378 with perfusion versus 668 without perfusion). Within the late window, patients with perfusion imaging had a shift towards better functional outcome at discharge compared with those without perfusion imaging (adjusted common odds ratio [OR], 1.45 [95% CI, 1.16–1.83]; P =0.001). There was no significant difference in functional independence (29.3% with perfusion versus 24.8% without; P =0.210) or in the safety outcome measures of symptomatic intracerebral hemorrhage ( P =0.53) and in-hospital mortality (10.6% with perfusion versus 14.3% without; P =0.053). In the early time window, patients with perfusion imaging had significantly improved odds of functional outcome (adjusted common OR, 1.51 [95% CI, 1.28–1.78]; P =0.0001) and functional independence (41.6% versus 33.6%, adjusted OR, 1.31 [95% CI, 1.08–1.59]; P =0.006). Perfusion imaging was associated with lower odds of futile recanalization in both time windows (late: adjusted OR, 0.70 [95% CI, 0.50–0.97]; P =0.034; early: adjusted OR, 0.80 [95% CI, 0.65–0.99]; P =0.047). Conclusions: In this real-world study, acquisition of perfusion imaging for EVT was associated with improvement in functional disability in the early and late time windows compared with nonperfusion neuroimaging. These indirect comparisons should be interpreted with caution while awaiting confirmatory data from prospective randomized trials.