Introduction L'apprentissage de la neuroanatomie est difficile en raison du manque de ressources interactives adaptées à la pratique neuroradiologique. Les supports d'apprentissage basé sur les jeux ont montré leur pertinence pour l'apprentissage de l’anatomie [1]. Notre objectif était de développer un outil interactif d'apprentissage de la neuroanatomie basé sur des atlas cérébraux de référence, et d'évaluer son acceptation au sein d'un service de neuroradiologie. Matériels et méthodes Nous avons créé neuroguessr.org, une plateforme web open-source basée sur la librairie niivue[1] et intégrant 15 atlas cérébraux de référence (Figure 1). Le site propose deux formats de jeu : des parties individuelles en mode apprentissage ou en mode défi (temps limité, tentatives restreintes), et des parties multijoueur où les joueurs s'affrontent pour localiser des régions anatomiques (Figure 2). Le site a été présenté aux internes et aux praticiens en imagerie diagnostique de notre service. Les statistiques d'utilisation anonymes ont été analysées sur 5 mois. Résultats Sur cette période, 985 parties individuelles et 59 parties multijoueur ont été jouées, totalisant 2700 tentatives d'identification de régions cérébrales. Les atlas les plus utilisés en solo ont été l'atlas des régions corticales (315/985, 32%), l'atlas des sous-champs hippocampiques (198/985, 20%) et l'atlas des territoires artériels (57/985, 6%). La majorité des parties solo se sont déroulées en mode apprentissage (734/985, 75%) plutôt qu'en mode défi (251/985, 25%). Les parties multijoueur ont rassemblé le plus souvent 2 joueurs (20/59, 34%), 5 joueurs (12/59, soit 20%) ou 3 joueurs (10/59, 17%). Conclusion Neuroguessr.org a rencontré une forte adhésion au sein de notre service de neuroradiologie. Une évaluation prospective actuellement en cours permettra de mesurer l'impact pédagogique de cette plateforme. Un défi quotidien sera proposé du 11 au 13 mars 2026, afin de permettre aux participants du congrès de la SFNR de découvrir l’application et de challenger leurs performances.
Introduction Anorexia nervosa (AN) often persists for years resulting in high morbidity and mortality rates. Hypoglycaemia, usually assessed from a single blood sample in the morning, is a critical severity indicator, continuous glucose monitoring (CGM) can therefore provide valuable data. Objectives This study aims to provide new glycaemic characteristics and look for their drivers, such as metabolic (present Body Mass Index, BMI) and duration of illness, in a sample of patients tested in real-world settings. Methods This cross-sectional study involved female outpatients with restricting subtype AN. Participants underwent CGM over 5 days in their usual environment. Data collected included age, BMI, illness duration, Eating Disorder Inventory-2 (EDI-2) scores. Glycaemic biomarkers such as hypoglycaemic area under the curve (AUC), mean and minimal glycaemia, and coefficient of variation were computed. Results Three hundred female patients were recorded for an average of 4.8 days. No significant correlation was found between glycaemic biomarkers and BMI. Illness duration significantly correlated with mean and minimum glycaemia (r = 0.26 and 0.23, respectively, p < 0.001) and hypoglycaemia AUC (r = -0.25, p < 0.001). Prolonged illness duration was associated with lower hypoglycaemia AUC and higher mean and minimal glycaemia. Conclusions In female patients with restricting subtype AN, illness duration rather than BMI may significantly impact glycaemic profiles. These results may suggest that glycaemic adaptations could be a factor that maintains undernutrition. AN should once more be conceptualized as a metabolic-psychiatric disorder. Disclosure of Interest None Declared
Background and Purpose: Persistent hypoperfusion despite successful endovascular treatment (EVT) for acute ischemic stroke (AIS) due to large-vessel occlusion (LVO), which results from either residual distal macrovascular occlusion or microvascular dysfunction ("no-reflow"), is increasingly recognized as a potential contributor to poor outcome. We aimed to characterize early post-EVT perfusion abnormalities on MRI and assessed their relationship with a panel of inflammatory biomarkers. Methods: We prospectively included patients with LVO-related AIS and successful (modified Thrombolysis in Cerebral Infarction [mTICI] ≥2b) recanalization who underwent MRI perfusion (MRP) as early as feasible after EVT and blood sampling before, during, immediately after and 24 hours after EVT. mTICI was assessed by an independent reader. Hypoperfusion within the affected vascular territory was assessed using three complementary approaches: (1) significant macrovascular hypoperfusion (Tmax>6s) in any location; (2) wedge-shaped perfusion deficits on Tmax maps, indicative of distal emboli; and (3) visual microvascular hypoperfusion within the infarct core on cerebral blood volume (CBV) and/or flow (CBF) maps (two-rater assessment). Associations between macrovascular hypoperfusion and 37 inflammatory biomarkers, measured using multiplex immunoassays, were analyzed using Mann–Whitney U tests with false discovery rate (FDR) correction. Results: Seventy-one patients were included. Median time from recanalization to MRP was 45 (19-118) min. Tmax>6s hypoperfusions were found in 21% of all patients; percentages decreased with better mTICI grades: 45%, 13% and 0% in mTICI 2b, 2c, and 3, respectively. Wedge-shaped hypoperfusion deficits were observed in 97% of mTICI2b, 44% of mTICI2c, and 8% mTICI3 cases (see Figure 1 for examples). Microvascular hypoperfusion was rare (6%, see Figure 2 for examples). Several inflammatory markers showed uncorrected associations with Tmax>6s hypoperfusion, but none survived FDR correction (Figure 3). Conclusions: Early perfusion deficits after EVT most often reflect residual distal emboli. Hypoperfusion on Tmax>6s maps was not associated with inflammatory biomarkers. Nevertheless, our integrated imaging and biomarker strategy—including samples collected immediately distal to the clot during EVT—lays the foundation for future mechanistic and therapeutic studies targeting microvascular reperfusion failure.
Humans are characterized by changes in neuronal development compared to non-human primates (NHPs), and metabolic changes are implicated in these species' differences. However, most research has focused on changes in the neurons and neural progenitor cells themselves, but astrocytes are crucial metabolic cells in the brain. Here, we identified a delayed onset of gliogenesis in human versus NHP organoid models. iPSC-derived astrocytes also revealed distinct metabolic profiles: NHP astrocytes exhibit increased serine and glycine synthesis, whereas human astrocytes show elevated lactate secretion, suggesting metabolic reprogramming of astrocytes across primate evolution. Astrocyte secretomes also differed, with human astrocyte secretomes enriched for axon-growth proteins. We then assessed the impact of these different species’ astrocyte secretomes on neuronal development. Conditioned media from NHPs as compared to humans led to increased electrophysiological activity and altered transcriptomic signatures of neuronal development in human neurons. Chemical PHGDH inhibition is consistent with a partial contribution of astrocytic serine/glycine metabolism to these species-specific effects. Together, these findings reveal a metabolic and secretory divergence between human and NHP astrocytes linked to species-specific neuronal development.
Light pollution is a major anthropogenic environmental change and a significant threat to ecosystems. Among other detrimental effects on physiology, artificial light at night (ALAN) disrupts circadian rhythms in a wide range of species. However, the underlying neuronal and genetic mechanisms remain poorly understood. Here, we show in Drosophila that the loss of the circadian clock gene period exacerbates the ALAN-induced shift towards nocturnal behaviour, with a more pronounced effect on males. In contrast, the loss of cycle has no such effect on males or females; cyc null mutants are nocturnal under standard light‒dark cycles, and their activity and sleep profiles are minimally or not affected by ALAN exposure. CRISPR-Cas9 knockout of period in most clock neurons resembles the null mutant phenotype. Our results show that mutations in components of the positive and negative limbs of the circadian clock result in distinct responses to nocturnal light and highlight the role of genetic factors in modulating behavioural plasticity in response to environmental perturbations.