Background: Emergency department (ED) revisits are critical quality indicators, particularly in medically underserved areas, where traditional prediction tools show limited performance. Machine learning (ML) approaches may offer improved predictive performance for identifying high-risk patients. Objective: This formative study aimed to develop and validate an ML-based model for predicting 30-day ED revisits using electronic health records from a tertiary hospital serving a medically underserved area in South Korea and to evaluate its clinical utility through interpretability analysis and risk stratification. Methods: This retrospective cohort study analyzed 36,230 adult patients visiting the Gangneung Asan Hospital ED in 2023. We developed and compared 3 ML models (extreme gradient boosting [XGBoost], random forest, and ElasticNet) using electronic health records. Model interpretability was ensured through Shapley additive explanations (SHAP) analysis, and clinical utility was evaluated through 5-tier risk stratification. Results: Among 36,230 patients, 798 (2.2%) revisited within 30 days. XGBoost achieved superior performance with an area under the receiver operating characteristic curve of 0.90 (95% CI 0.88‐0.92), a sensitivity of 0.94, and a specificity of 0.69. The SHAP analysis identified ED length of stay, oxygen saturation, systolic blood pressure, computed tomography performance, antibiotic use, and liver disease as key predictors. Risk stratification demonstrated a 25-fold difference in the actual revisit rates between the lowest (152/8450, 1.8%) and the highest (686/1500, 45.7%) risk groups. Conclusions: The XGBoost model demonstrated excellent predictive performance with high interpretability for 30-day ED revisit predictions. The implementation of this model could enable risk-stratified interventions and more efficient resource allocation in medically underserved settings, potentially reducing unnecessary revisits and improving patient outcomes. This formative study establishes feasibility and provides a foundation for future multicenter validation studies in similar medically underserved settings.
<img src="https://jmir-production.s3.us-east-2.amazonaws.com/thumbs/b4d0c970ac5696468dd8cdba2c4c476a" />
Neuronal Protein Tracing Reveals How the Brain Routes Its Waste
The brain is one of the busiest organs in the body, constantly processing and reshaping itself. That activity produces an equally constant stream of molecular byproducts—proteins that need to be moved out before they accumulate. When those clearance routes slow or break down, waste lingers, and the consequences can be profound. In Alzheimer’s disease, for example, toxic proteins build up in vulnerable regions. Yet despite decades of research, scientists have lacked a clear view of how waste normally leaves the brain.
A new study from the Gladstone Institutes offers the clearest picture yet of how the brain normally takes out its trash—and what happens when those routes fail. Published in Cell as “Physiological brain clearance architecture revealed by neuronal protein tracing,” the work introduces a method that traces waste proteins from the moment they are produced inside neurons to the moment they leave the brain.
For decades, researchers have relied on injecting tracers into the cerebrospinal fluid (CSF) to visualize drainage. But this approach, while illuminating, shows all possible routes, instead of pinpointing the most-used exit. “These injected tracers disturb the very system we’re attempting to measure,” said lead author Andrew Yang, PhD, a Gladstone investigator. “We wanted to find a better way.”
Yang’s team engineered neurons in mice to produce a fluorescent protein, ZsGreen, that could be followed as it exited the brain through its natural routes. This allowed the researchers to track waste as it moved into the dura, skull, nasal cavity, and lymph nodes—regions populated by specialized immune cells that interact with brain‑derived proteins.
The resulting map diverged sharply from the field’s long‑held assumptions. Traditional CSF tracers had pointed to the cervical lymph nodes as a major drainage site. But the new method revealed that very little neuronal waste actually reaches those nodes. “We were surprised to find that very little ZsGreen drained to the cervical lymph nodes,” Yang said. “Instead, waste drained through the dura, skull, and nasal cavity. Our findings underscore why tracking waste proteins themselves, rather than movement of the cerebrospinal fluid, provides a more accurate understanding of waste clearance dynamics.”
The team also uncovered a striking organizational principle: where a protein is made determines where it drains. Proteins produced in upper forebrain regions exited through upper routes, while those from deeper structures, such as the striatum, used lower pathways. The researchers call this the brain’s “nearest‑exit” model. “It’s like each brain region has a biological ZIP code system to ensure waste will be sent to the correct drainage site,” said Nalini Rao, PhD, a postdoctoral fellow. She noted that in aging or disease, these ZIP codes may become scrambled, potentially explaining why certain regions are more vulnerable to disorders like Alzheimer’s.
Disease models reinforced the system’s fragility. In mice with acute inflammation, ZsGreen leaked directly into the bloodstream, bypassing normal routes. In an Alzheimer’s model, waste became trapped inside the brain, unable to drain effectively. “Understanding how diseases disrupt brain clearance could help us design therapeutics to target the brain border compartments and enhance waste removal,” Rao said.
With their new tracing method, Yang’s group plans to probe how clearance changes across aging, sleep, and disease—and whether brain tumors exploit these pathways to evade immune detection. The architecture of brain waste disposal, once opaque, is now open for exploration.
The post Neuronal Protein Tracing Reveals How the Brain Routes Its Waste appeared first on GEN – Genetic Engineering and Biotechnology News.
Vascular endothelial growth factor receptor-1 (VEGFR-1) knock-down is protective against hypoxia, Aβ1-42 oligomer and Aβ1-42 fibril -induced neuronal cell death: implications in AD pathogenesis
IntroductionRecent transcriptome analysis has demonstrated increased expression of Vascular Endothelial Growth Factor receptor-1 (VEGFR-1/FLT1) and in AD brain. Increased expression of VEGFR1 and its ligand VEGFB were associated with a more rapid rate of cognitive decline, providing evidence of a potential link between increased VEGFR-1 expression in AD pathogenesis. In this study, we explored the potential role of VEGFR-1 expression in neurons on AD pathology.MethodsTo confirm VEGFR1 expression in AD brains, we first performed immunostaining in AD brain sections (AD – Braak stage V-VI, and normal controls – Braak 0-II). And to determine a potential detrimental role of neuronal VEGFR1 expression on AD associated pathologies, we exposed SH-SY5Y human neuroblastoma cells and mouse primary neurons to either hypoxia conditions (1%O2) or 5 μ Aβ1-42 oligomers or fibrils for 24, 28 and 72hrs.ResultsIn this study, we found preferential staining of VEGFR-1 in the neuropil and neuronal cell bodies both in AD and Control hippocampus and increased VEGFR-1 immunoreactivity in dystrophic neuritic processes in the vicinity of Thio-S positive amyloid plaques in AD brains. And treatment of SH-SY5Y human neuroblastoma cell line and mouse primary neurons, with either hypoxia conditions or Aβ1-42 oligomers, resulted in increased VEGFR-1 expression and cleaved caspase 3 activation, leading to neuronal toxicities/cell death. Similarly, treatment with Aβ1-42 fibrils also increased VEGFR-1 and cleaved caspase 3 protein levels in the SH-SY5Y cells whereas treatment with Aβ1-42 monomers had no effect on VEGFR-1 expression. In addition, we show that over-expression of VEGFR-1 intracellular domains in SH-SY5Y cells directly induced neuronal toxicities and importantly, siRNA-mediated knockdown of VEGFR-1 in neurons prevented the hypoxia, Aβ1-42 oligomer and Aβ1-42 fibril-induced toxicities and cell death phenotypes. Treatment with either hypoxia or Aβ1-42 oligomers also reduced expression of cell survival genes including VEGFR-2 and Hippo pathway YAP1 and siRNA-mediated VEGFR-1 knockdown in the neurons normalized expression of both VEGFR-2 and YAP1. Using differential gene expression analysis, we demonstrated upregulation of several inflammatory/interferon-stimulated genes (ISGs) as well as increased expression of genes involved in activation of oxidative stress and cell death pathways in response to Aβ1-42 oligomers treatment in mouse primary neurons. And siRNA-mediated VEGFR-1 knockdown in the mouse primary neurons, reduced gene expression of both the ISGs and oxidative stress/cell death pathways in response to Aβ1-42 oligomer treatment.DiscussionIn summary, these results show that siRNA-mediated knockdown of VEGFR-1 in neurons significantly prevented hypoxia, Aβ1-42 oligomer and Aβ1-42 fibril-induced cellular toxicities and cell death phenotypes, indicating a potential detrimental role of aberrant VEGFR-1 expression and signaling in response to AD associated pathologies.
Pain-side-specific alteration of structural networks in trigeminal neuralgia: a connectome analysis
ObjectivesTrigeminal neuralgia (TN) involves disruption in the integrity of the white matter, the side-specific pain topology of these alterations at the network has yet to be defined. In this study, we investigated the lateralization of structural network architecture and nodal characteristics in TN patients.MethodsWhole-brain structural networks (90 × 90 connectivity matrices) were reconstructed from diffusion tensor imaging (DTI) tractography data of 30 TN patients and 20 matched controls. We applied Network-Based Statistics (NBS) to detect altered connectivity sub-networks, and graph theoretical analysis to profile global and nodal properties. Our analysis aimed to delineate changes that were specific to the painful side.ResultsNBS analysis revealed that structural connectivity formed subnetworks involving multiple functional networks. A subnetwork involving the anterior cingulate gyrus (ACG) and postcentral gyrus (S1) was identified on the painful side, indicating that TN stimulation may enhance structural connectivity between regions related to salience and somatosensory processing, thereby facilitating the acceleration of pain perception and response. On the non-pain side, we observed enhanced structural connections between visual and attention-related regions. The third subnetwork was characterized by widespread and non-focal reductions in fiber tract connectivity. However, despite these localized alterations, the global network properties of the brain in TN patients remained stable, with node-specific properties undergoing alterations in multiple brain regions, including the cuneus, inferior parietal lobule, and superior frontal gyrus.ConclusionHerein, we applied NBS and graph theoretical analysis to investigate changes in the structural brain networks of patients with TN. Analysis revealed that specific subnetworks and key nodes can be affected by TN. We also confirmed obvious differences in the involved subnetworks between pain and non-pain sides in TN patients. These findings suggest that these specific subnetworks and nodes could represent valuable biomarkers for clinical evaluation and intervention in TN patients.
Emotion recognition based on the temporal patterns of electroencephalogram signals and electrodermal response signals using the TRANSFORMER network
IntroductionEmotion recognition using physiological signals plays an important role in affective neuroscience and human-centered artificial intelligence. Current methods still face challenges in long-range temporal dependency modeling and explicit central–autonomic coupling representation, while generalization under subject-independent protocols needs further improvement.MethodsThis study proposes a Transformer-based multimodal framework for four-class discrete emotion recognition (neutral, happiness, sadness, and fear) by jointly modeling EEG and GSR signals. The architecture integrates temporal self-attention and bidirectional cross-modal attention. Experiments were conducted on 42 neurologically healthy adults with a controlled audiovisual emotion elicitation paradigm, evaluated using subject-independent five-fold cross-validation.ResultsThe model achieved a mean classification accuracy of 87.42% ± 2.13%, with precision of 87.6%, recall of 87.4%, and F1-score of 87.5%. It outperformed CNN and Bi-LSTM baselines by 4.91% and 6.38%, respectively. Multimodal fusion significantly boosted high-arousal emotion recognition, with fear accuracy increasing from 82.11% (EEG-only) to 88.63% (p = 0.004).DiscussionThese findings confirm that long-range temporal modeling and explicit cross-modal interaction can substantially improve multimodal physiological emotion recognition. The proposed framework is scalable and interpretable, advances central–autonomic coupling modeling, enhances generalization via strict subject-independent validation, and supports physiological interpretability through attention visualization and modality sensitivity analysis.
Leveraging Synchrosqueezing Transform (SST)-based representations in a dual-stream attention framework to enhance sleep apnea detection and subtyping
IntroductionSleep apnea and hypopnea syndrome (SAHS) is a prevalent disorder with profound adverse effects on health and overall quality of life, thereby necessitating the development of accurate and accessible screening tools. Electrocardiogram (ECG)-based analysis, being non-invasive and readily deployable in low-cost hardware, offers a particularly convenient approach for SAHS screening and preliminary diagnosis. However, conventional time-frequency analysis often fails to capture the subtle yet critical patterns in ECG signals due to the Heisenberg uncertainty principle, leading to limited resolution and information loss.MethodsTo overcome these limitations, this study proposes a Dual Stream Cross Attention Fusion Network (DSCAFNet) based on the uncertainty-mitigated time-frequency representations generated via Synchrosqueezing Transform (SST). The framework uniquely constructs two complementary, high-fidelity SST-based representations, which are strategically designed to provide distinct yet synergistic perspectives on the complex, non-stationary dynamics of SAHS. A dedicated cross-attention fusion module then harnesses these complementary views, enabling the model to discriminatively integrate multi-resolution features for significantly enhanced pattern recognition.ResultsExtensively evaluated on the public Apnea-ECG dataset, DSCAFNet achieves an accuracy of 0.9572, a sensitivity of 0.9575, a specificity of 0.9584, and an F1-score of 0.9557, performing on par with state-of-the-art methods. More importantly, rigorous validation on a private Huashan-apnea dataset yields an accuracy of 0.9003 for binary classification and 0.7564 for four-class subtyping, demonstrating strong effectiveness and generalization.ConclusionThese consistent results across datasets highlight DSCAFNet as a promising framework for intelligent and accessible SAHS screening, with potential for integration into portable data acquisition systems combined with cloud-based analysis.
The Sentinel Phenotype: a theoretical bioenergetic and neurobiological framework for high-fidelity predictive systems (HEPOE Theory)
The HEPOE Theory (High Entropy Predictive Organization Efficiency) proposes a novel conceptual framework for understanding Giftedness (HA/G), moving beyond academic performance-based models toward a biophysical and neuroscientific foundation. Through a theoretical synthesis grounded in the Free Energy Principle and Biological Thermodynamics, the gifted individual is redefined as a “Sentinel”: a high-fidelity sampling system specialized in the early detection of isomorphy and the reduction of systemic entropy. This framework reinterprets Charles Spearman’s general intelligence (g) as a macroscopic manifestation of hardware efficiency, where reasoning ability is proposed to be fundamentally constrained by working-memory capacity and the metabolic economy of ATP resynthesis. We hypothesize that the hardware operates under an “open sensory gating” regime and low latent inhibition, leading to high metabolic costs and chronic allostatic load. The paper introduces the original concept of Predictive Moral Injury to conceptualize the potential somatic damage resulting from the early perception of ethical-systemic collapses within low-resolution environments. The HEPOE Unification Matrix integrates decades of classical literature and proposes a rigorous differential diagnosis against the pathologization of ASD, ADHD, and PTSD. It hypothesizes that the Sentinel’s exhaustion is not a dysfunction, but a logistical byproduct of high predictive performance under entropy-saturated conditions.
Perceptions of mental health services among informal caregivers in Sardinia, Italy, during the post-pandemic crisis of the national health system: a comparative study between 2024 and 2025
BackgroundItaly’s community-based mental health model, grounded in Law 180, has long been regarded as a global example of rights-oriented psychiatric reform. However, recent years have witnessed a crisis in the national health system, marked by chronic and progressively accumulating underfunding, workforce shortages, and organizational strain, rather than a single acute disruption. This s/tudy compares two samples of informal caregivers of individuals receiving mental health care in Sardinia (2024 vs. 2025) to explore differences in their perceptions of service quality, organizational well-being, and respect for human rights.MethodsA cross-sectional design was used to compare independent caregiver samples (2024: n = 100; 2025: n = 74). Participants completed the Well-Being at Work and Respect for Human Rights Questionnaire (WWRR), which assesses satisfaction with care, perceptions of organizational quality, and respect for users’ and staff’s rights. Data were analyzed with ANOVA and chi-square tests.ResultsCaregivers in 2025 reported significantly higher satisfaction than those in 2024 regarding the quality of services (p = 0.007), organizational aspects (p = 0.007), and respect for the human rights of both users (p < 0.001) and staff (p = 0.037). No differences emerged in perceived user satisfaction or resource adequacy, with both cohorts expressing persistent concern about staffing and infrastructure. The perceived need for additional nurses, doctors, and support staff increased in 2025, which may indicate growing awareness of workforce fragility.ConclusionsThese differences in satisfaction should be interpreted within a context of prolonged systemic strain, which may foster adaptive or resilience-based perceptions among informal caregivers. Despite ongoing structural difficulties, informal caregivers maintain high confidence in Italy’s community mental health services and perceive them as respectful of human rights. However, their increasing concern about resource shortages highlights the urgent need for investment to preserve the ethical and organizational strengths of the Italian model.
Hyponatremia in patients with severe anorexia nervosa was associated with more severe and longer duration of disease
IntroductionSeveral mechanisms are thought to contribute to hyponatremia in patients with anorexia nervosa (AN). The aims of this descriptive, cross-sectional study among patients admitted to a specialized somatic unit for eating disorders (ED) were to determine the frequency of hyponatremia and to compare medical findings between patients with normonatremia and hyponatremia.MethodsThis retrospective, descriptive cross-sectional study included patients admitted to the unit between December 2016 and October 2021. Demographic, medical history, and clinical data were extracted from medical records. Patients were categorized according to plasma sodium concentration (<135 mmol/L vs. ≥135 mmol/L).ResultsAmong 131 patients, 17 (13%) had hyponatremia at admission. Hyponatremia was associated with lower BMI, lower nadir BMI, longer disease duration, and an adverse biochemical profile (lower albumin, higher creatinine, higher platelet counts, and higher bicarbonate levels). Thirteen patients (10%) were deceased at follow-up; hyponatremia was associated with mortality in unadjusted analysis (OR 8.03, 95% CI 2.29–28.16) but not after multivariable adjustment.DiscussionThe study found that 13% of patients admitted to a specialized somatic unit for ED had hyponatremia, which clustered with indicators of more severe and longstanding AN (lower BMI, lower nadir BMI, longer disease duration). Hyponatremia was associated with mortality in unadjusted analyses, but this association was attenuated after adjustment for age and illness severity. Despite the established potential of purging to induce hyponatremia, our findings suggest that, in this severely ill inpatient population, overall illness severity and chronic medical deterioration may be more important determinants of both hyponatremia and mortality risk than purging per se.
Metacognitive model of suicidality: a study of Iranian inpatients
BackgroundThe metacognitive model of suicidality proposes that positive metacognitions about suicide activate suicide-specific rumination, which in turn leads to the activation of negative metacognitions about suicide and an escalating aggravation of suicidal ideation/behavior. Initial studies support the model assumptions. However, investigations in highly burdened inpatient samples as well as studies in non-Western samples are missing by now.MethodsA total of 209 Iranian psychiatric inpatients (56.9% female; age M = 31.14, SD = 11.04) took part in a cross-sectional assessment. Self-report measures to assess suicidal ideation/behavior, depressive symptoms, suicide-specific rumination, and metacognitions about suicide were used.ResultsPositive metacognitions about suicide were associated with suicide-specific rumination. Suicide-specific rumination was associated with negative metacognitions about suicide. Suicidal thoughts were positively and positive metacognitions about suicide were negatively associated with lifetime suicide attempts.ConclusionThe results support assumptions of the metacognitive model of suicidality and underscore the importance of metacognitions about suicide and suicide-specific rumination in understanding the suicidal process.