Hasil untuk "Neurosciences. Biological psychiatry. Neuropsychiatry"

Yangyang Guo, Bingyang Zhang, Lianmei Zhong et al.

BackgroundIntracranial and/or extracranial atherosclerotic stenosis is a common etiology of acute ischemic stroke (AIS). This study aimed to evaluate the impact of intracranial or extracranial atherosclerotic stenosis on early neurological deterioration (END), hemorrhagic transformation (HT) and 90-day clinical outcomes in patients receiving intravenous thrombolysis.MethodsWe retrospectively enrolled patients with AIS who received intravenous alteplase (0.9 mg/kg) at the First Affiliated Hospital of Kunming Medical University between February 2019 and August 2022. Data on demographics, stroke risk factors, laboratory results, and neuroimaging findings were collected. Atherosclerotic stenosis (AS) was defined as >50% intracranial or extracranial arteries. Logistic regression was performed to identify independent predictors of clinical outcomes. END was defined as an increase of ≥4 points in the National Institutes of Health Stroke Scale (NIHSS) score within 24 h after stroke onset. HT was defined as any newly detected intracranial hemorrhage on follow-up cranial CT performed within 7 days after symptom onset.ResultsA total of 185 AIS patients receiving intravenous thrombolysis were included in this study, with 88 (47.6%) in the IEAS group and 97 (52.4%) in the non-stenosis group. There was no significant association between the incidence of END and the presence of IEAS. Multivariable regression analysis revealed that baseline NIHSS was an independent risk factor for HT (OR = 1.120, 95% CI 1.038–1.209, p = 0.003), 90-day poor clinical outcome (OR = 1.198, 95% CI 1.105–1.298, p = 0.001) and 90-day death (OR = 1.384, 95% CI 1.179–1.625, p = 0.001). Although IEAS was not significantly associated with the incidence of END or HT, it was significantly correlated with 90-day poor clinical outcome (OR = 1.350, 95% CI 1.108–1.644, p = 0.003).ConclusionsIn this cohort, IEAS was not associated with END or HT but emerged as an independent predictor of poor 90-day functional outcome after intravenous thrombolysis for AIS.

Yifan Wu, Jiyue Jiang, Xichen Ye et al.

Biological foundation models (BioFMs), pretrained on large-scale biological sequences, have recently shown strong potential in providing meaningful representations for diverse downstream bioinformatics tasks. However, such models often rely on millions to billions of training sequences and billions of parameters, resulting in prohibitive computational costs and significant barriers to reproducibility and accessibility, particularly for academic labs. To address these challenges, we investigate the feasibility of data pruning for BioFM pretraining and propose a post-hoc influence-guided data pruning framework tailored to biological domains. Our approach introduces a subset-based self-influence formulation that enables efficient estimation of sample importance at low computational cost, and builds upon it two simple yet effective selection strategies, namely Top-k Influence (Top I) and Coverage-Centric Influence (CCI). We empirically validate our method on two representative BioFMs, RNA-FM and ESM-C. For RNA, our framework consistently outperforms random selection baselines under an extreme pruning rate of over 99 percent, demonstrating its effectiveness. Furthermore, we show the generalizability of our framework on protein-related tasks using ESM-C. In particular, our coreset even outperforms random subsets that are ten times larger in both RNA and protein settings, revealing substantial redundancy in biological sequence datasets. These findings underscore the potential of influence-guided data pruning to substantially reduce the computational cost of BioFM pretraining, paving the way for more efficient, accessible, and sustainable biological AI research.

Noriaki Sato, Marco Scutari, Shuichi Kawano et al.

Estimating causal networks from biological data is a critical step in systems biology. When evaluating the inferred network, assessing the networks based on their intervention effects is particularly important for downstream probabilistic reasoning and the identification of potential drug targets. In the context of gene regulatory network inference, biological databases are often used as reference sources. These databases typically describe relationships in a qualitative rather than quantitative manner. However, few evaluation metrics have been developed that take this qualitative nature into account. To address this, we developed a metric, the sign-augmented Structural Intervention Distance (sSID), and a weighted sSID that incorporates the net effects of the intervention. Through simulations and analyses of real transcriptomic datasets, we found that our proposed metrics could identify a different algorithm as optimal compared to conventional metrics, and the network selected by sSID had a superior performance in the classification task of clinical covariates using transcriptomic data. This suggests that sSID can distinguish networks that are structurally correct but functionally incorrect, highlighting its potential as a more biologically meaningful and practical evaluation metric.

Toshiki Tazoe

Chi-Yang Hsu, Kyle Cox, Jiawei Xu et al.

We present the Thought Graph as a novel framework to support complex reasoning and use gene set analysis as an example to uncover semantic relationships between biological processes. Our framework stands out for its ability to provide a deeper understanding of gene sets, significantly surpassing GSEA by 40.28% and LLM baselines by 5.38% based on cosine similarity to human annotations. Our analysis further provides insights into future directions of biological processes naming, and implications for bioinformatics and precision medicine.

Nima Dehghani, Michael Levin

The pursuit of creating artificial intelligence (AI) mirrors our longstanding fascination with understanding our own intelligence. From the myths of Talos to Aristotelian logic and Heron's inventions, we have sought to replicate the marvels of the mind. While recent advances in AI hold promise, singular approaches often fall short in capturing the essence of intelligence. This paper explores how fundamental principles from biological computation--particularly context-dependent, hierarchical information processing, trial-and-error heuristics, and multi-scale organization--can guide the design of truly intelligent systems. By examining the nuanced mechanisms of biological intelligence, such as top-down causality and adaptive interaction with the environment, we aim to illuminate potential limitations in artificial constructs. Our goal is to provide a framework inspired by biological systems for designing more adaptable and robust artificial intelligent systems.

E. Díaz-Mesa, C. Cárdenes Moreno, A. Morera-Fumero et al.

Introduction Malondialdehyde (MDA) is a product of polyunsaturated fatty acid peroxidation (Del Rio D, et al. A review of recent studies on MDA as toxic molecule and biological marker of oxidative stress. Nutr Metab Cardiovasc Dis. 2005;15:316-28). It is a biomarker of oxidative stress and is involved in the pathophysiology of schizophrenia (Goh et al. Asian J Psychiatr. 2022;67:102932). Schizofrenia is linked to disrupted oxidative balance and inflammation (Więdłocha et al. Brain Sci. 2023;13:490). Prior research has shown connections between biomarkers and circadian rhythms in schizophrenia (Morera & Abreu. Acta Physiol Scand. 2007;43:313-14) and diabetes type 2 (Kanabrocki EL, et al. Circadian variation in oxidative stress biomarkers in healthy and type II diabetic men. Chronobiol Int. 2002;19:423-39). To determinate if MDA levels have a role in schizophrenia and follow a circadian rhythm may be useful. Objectives The aim of our study is to compare diurnal and nocturnal MDA serum levels in patients in acute and stabilized phases of schizophrenia according to CIE-10 to find out if there are variations related with circadian rhythms Methods 47 patients were included in our study in two clinical phases: acute episode and stabilization. Blood samples were collected at 12:00h and at 00:00h. MDA serum levels were measured twice: when patients were decompensated (admission) and at clinical stabilization (discharge). The relationship between quantitative variables at both times was analysed by T-Student test Results There is no significative difference between night and day MDA levels in the acute phase of the schizophrenia (2.22±1.352 vs. 1.93±1.530, p<0.09). There is statistical significance between 12:00 and 00:00 (1.90±1.136 vs. 1.34±0.868, p<0.001) at discharge: it was observed that levels decreased. This result can be interpreted as there is circadian rhythm in stabilized phases. Conclusions MDA levels in patients with schizophrenia do not follow a circadian rhythm in the acute episode. When they are clinically stabilized present a circadian change. These patients lose the circadian rhythm in acute episodes. MDA circadian rhythm may help diagnose the clinical phase and its severity. It is necessary to perform more studies to know its utility as an oxidative biomarker Disclosure of Interest None Declared

Aaron D. Talsma, Jon P. Niemi, Richard E. Zigmond

Abstract Background Since the 1990s, evidence has accumulated that macrophages promote peripheral nerve regeneration and are required for enhancing regeneration in the conditioning lesion (CL) response. After a sciatic nerve injury, macrophages accumulate in the injury site, the nerve distal to that site, and the axotomized dorsal root ganglia (DRGs). In the peripheral nervous system, as in other tissues, the macrophage response is derived from both resident macrophages and recruited monocyte-derived macrophages (MDMs). Unresolved questions are: at which sites do macrophages enhance nerve regeneration, and is a particular population needed. Methods Ccr2 knock-out (KO) and Ccr2 gfp/gfp knock-in/KO mice were used to prevent MDM recruitment. Using these strains in a sciatic CL paradigm, we examined the necessity of MDMs and residents for CL-enhanced regeneration in vivo and characterized injury-induced nerve inflammation. CL paradigm variants, including the addition of pharmacological macrophage depletion methods, tested the role of various macrophage populations in initiating or sustaining the CL response. In vivo regeneration, measured from bilateral proximal test lesions (TLs) after 2 d, and macrophages were quantified by immunofluorescent staining. Results Peripheral CL-enhanced regeneration was equivalent between crush and transection CLs and was sustained for 28 days in both Ccr2 KO and WT mice despite MDM depletion. Similarly, the central CL response measured in dorsal roots was unchanged in Ccr2 KO mice. Macrophages at both the TL and CL, but not between them, stained for the pro-regenerative marker, arginase 1. TL macrophages were primarily CCR2-dependent MDMs and nearly absent in Ccr2 KO and Ccr2 gfp/gfp KO mice. However, there were only slightly fewer Arg1+ macrophages in CCR2 null CLs than controls due to resident macrophage compensation. Zymosan injection into an intact WT sciatic nerve recruited Arg1+ macrophages but did not enhance regeneration. Finally, clodronate injection into Ccr2 gfp KO CLs dramatically reduced CL macrophages. Combined with the Ccr2 gfp KO background, depleting MDMs and TL macrophages, and a transection CL, physically removing the distal nerve environment, nearly all macrophages in the nerve were removed, yet CL-enhanced regeneration was not impaired. Conclusions Macrophages in the sciatic nerve are neither necessary nor sufficient to produce a CL response.

Chenquan Lin, Shuangyang Zhang, Ping Yang et al.

Abstract The prolonged usage of atypical antipsychotic drugs (AAPD) among individuals with schizophrenia often leads to metabolic side effects such as dyslipidemia. These effects not only limit one’s selection of AAPD but also significantly reduce compliance and quality of life of patients. Recent studies suggest that bilirubin plays a crucial role in maintaining lipid homeostasis and may be a potential pre-treatment biomarker for individuals with dyslipidemia. The present study included 644 schizophrenia patients from two centers. Demographic and clinical characteristics were collected at baseline and 4 weeks after admission to investigate the correlation between metabolites, episodes, usage of AAPDs, and occurrence of dyslipidemia. Besides, we explored the combined predictive value of genotypes and baseline bilirubin for dyslipidemia by employing multiple PCR targeted capture techniques to sequence two pathways: bilirubin metabolism-related genes and lipid metabolism-related genes. Our results indicated that there existed a negative correlation between the changes in bilirubin levels and triglyceride (TG) levels in patients with schizophrenia. Among three types of bilirubin, direct bilirubin in the baseline (DBIL-bl) proved to be the most effective in predicting dyslipidemia in the ROC analysis (AUC = 0.627, p < 0.001). Furthermore, the odds ratio from multinomial logistic regression analysis showed that UGT1A1*6 was a protective factor for dyslipidemia (ß = −12.868, p < 0.001). The combination of baseline DBIL and UGT1A1*6 significantly improved the performance in predicting dyslipidemia (AUC = 0.939, p < 0.001). Schizophrenia patients with UGT1A1*6 mutation and a certain level of baseline bilirubin may be more resistant to dyslipidemia and have more selections for AAPD than other patients.

Christopher R. Bailie, Pooja S. Pillai, Atul Goodwin Singh et al.

There are few evidence-based interventions to support caregiver mental health developed for low- and middle-income countries. Nae Umeed is a community-based group intervention developed with collaboratively with local community health workers in Uttarakhand, India primarily to promote mental wellbeing for caregivers and others. This pre–post study aimed to evaluate whether Nae Umeed improved mental health and social participation for people with mental distress, including caregivers. The intervention consisted of 14 structured group sessions facilitated by community health workers. Among 115 adult participants, 20% were caregivers and 80% were people with disability and other vulnerable community members; 62% had no formal education and 92% were female. Substantial and statistically significant improvements occurred in validated psychometric measures for mental health (12-Item General Health Questionnaire, Patient Health Questionnaire-9) and social participation (Participation Scale). Improvements occurred regardless of caregiver status. This intervention addressed mental health and social participation for marginalised groups that are typically without access to formal mental health care and findings suggest Nae Umeed improved mental health and social participation; however, a controlled community trial would be required to prove causation. Community-based group interventions are a promising approach to improving the mental health of vulnerable groups in South Asia.

Taras O. Simaniv, Anna A. Belkina, Maria N. Zakharova

Demyelinating diseases of the central nervous system and multiple sclerosis in particular are a pressing issue for medical community and society as a whole. Deve- lopment and implementation of highly effective specific therapy significantly slow the disease progression and help maintain patients' quality of life and social participation. We analyzed pathogenic mechanisms of multiple sclerosis and other B cell-mediated diseases and reviewed therapeutic options for main disease stages.

Peter F. Liddle, Elizabeth B. Liddle

Current diagnostic criteria for schizophrenia place emphasis on delusions and hallucinations, whereas the classical descriptions of schizophrenia by Kraepelin and Bleuler emphasized disorganization and impoverishment of mental activity. Despite the availability of antipsychotic medication for treating delusions and hallucinations, many patients continue to experience persisting disability. Improving treatment requires a better understanding of the processes leading to persisting disability. We recently introduced the term classical schizophrenia to describe cases with disorganized and impoverished mental activity, cognitive impairment and predisposition to persisting disability. Recent evidence reveals that a polygenic score indicating risk for schizophrenia predicts severity of the features of classical schizophrenia: disorganization, and to a lesser extent, impoverishment of mental activity and cognitive impairment. Current understanding of brain function attributes a cardinal role to predictive coding: the process of generating models of the world that are successively updated in light of confirmation or contradiction by subsequent sensory information. It has been proposed that abnormalities of these predictive processes account for delusions and hallucinations. Here we examine the evidence provided by electrophysiology and fMRI indicating that imprecise predictive coding is the core pathological process in classical schizophrenia, accounting for disorganization, psychomotor poverty and cognitive impairment. Functional imaging reveals aberrant brain activity at network hubs engaged during encoding of predictions. We discuss the possibility that frequent prediction errors might promote excess release of the neurotransmitter, dopamine, thereby accounting for the occurrence of episodes of florid psychotic symptoms including delusions and hallucinations in classical schizophrenia. While the predictive coding hypotheses partially accounts for the time-course of classical schizophrenia, the overall body of evidence indicates that environmental factors also contribute. We discuss the evidence that chronic inflammation is a mechanism that might link diverse genetic and environmental etiological factors, and contribute to the proposed imprecision of predictive coding.

Ellen van der Plas, Jeffrey D. Long, Timothy R. Koscik et al.

IntroductionThe present study had four aims. First, neuronal injury markers, including neurofilament light (NF-L), total tau, glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase (UCH-L1), were compared between individuals with and without adult-onset myotonic dystrophy type 1 (DM1). Second, the impact of age and CTG repeat on brain injury markers was evaluated. Third, change in brain injury markers across the study period was quantified. Fourth, associations between brain injury markers and cerebral white matter (WM) fractional anisotropy (FA) were identified.MethodsYearly assessments, encompassing blood draws and diffusion tensor imaging on a 3T scanner, were conducted on three occasions. Neuronal injury markers were quantified using single molecule array (Simoa).ResultsThe sample included 53 patients and 70 controls. NF-L was higher in DM1 patients than controls, with individuals in the premanifest phases of DM1 (PreDM1) exhibiting intermediate levels (χ(2)2=38.142, P &lt; 0.001). Total tau was lower in DM1 patients than controls (Estimate = −0.62, 95% confidence interval [CI] −0.95: −0.28, P &lt; 0.001), while GFAP was elevated in PreDM1 only (Estimate = 30.37, 95% CI 10.56:50.19, P = 0.003). Plasma concentrations of UCH-L1 did not differ between groups. The age by CTG interaction predicted NF-L: patients with higher estimated progenitor allelege length (ePAL) had higher NF-L at a younger age, relative to patients with lower CTG repeat; however, the latter exhibited faster age-related change (Estimate = −0.0021, 95% CI −0.0042: −0.0001, P = 0.045). None of the markers changed substantially over the study period. Finally, cerebral WM FA was significantly associated with NF-L (Estimate = −42.86, 95% CI −82.70: −3.02, P = 0.035).InterpretationWhile NF-L appears sensitive to disease onset and severity, its utility as a marker of progression remains to be determined. The tau assay may have low sensitivity to tau pathology associated with DM1.

M. Rapado-Castro, C. Arango

Now more than ever, preventative and interventional strategies need to be put in place to overcome the collateral damage caused by the pandemic. Mental health is one of the casualties. A long time has passed since the Wuhan outbreak, and the world is or will be soon experiencing the aftermath of the first modern pandemic. Now is, therefore, the time to take notice of the effects of sustained stress on mental and physical health over time and to take action to address them. The uncertainty of the COVID-19 pandemic has been a challenge to mental health globally, transforming the way people make sense of the world and how they relate to one another. The pandemic has presented us with both physical and mental challenges, including the continued stress of lockdowns, restrictions, social distancing, job losses, income insecurity, health and social inequality, suffering, and isolation. Provision of services has changed [1], and the consequences of the pandemic for mental health have become particularly evident in vulnerable groups such as people with pre-existing mental health conditions and individuals at risk for mental disorders, children and adolescents in particular. The incidence of stress-related mental disorders has increased throughout the pandemic, with suicidal crises, anxiety disorders, eating disorders, and major depressive episodes been the most compelling in this population [2]. A recent survey to estimate the impact of the COVID-19 pandemic on child and adolescent psychiatry services in Europe [2] found a dramatic increase in referrals for assessment and hospitalizations 1 year into the pandemic, with perception of the impact of the pandemic on mental health and psychopathology in this population escalating from moderate in 2020 to extreme in 2021. In this context, lessons must be learnt about mechanisms of cope, resilience, and adaptation. As Chmitorz et al. [3] noted in this issue of European Archives of Psychiatry and Clinical Neuroscience, there is an urgent need to investigate protective mechanisms that support the maintenance of mental health during and after adversity. These authors focus on cognitive, physiological, and neural pathways (i.e., resilience mechanisms) that may provide protection against stress-related impairments in a large sample of 1191 individuals. They present baseline descriptive data from the ongoing Longitudinal Resilience Assessment (LORA) study, a population-based study aiming to characterize resilience as a continued process of biological, neural, and cognitive adaptation to stressors and daily hassles, directly assessing their influence on the human organism over a period of at least 3 years. It is very timely that these authors are using a comprehensive, granular approach to the investigation of the general (lifestyle, cortisol reactivity, microbiome, genetic, epigenetic and cognitive) mechanisms involved in resilience in adults from the community, performing an in-depth exploration of their mental health status, ability to recover from stress, major life events, daily hassles, and perceived stress mechanisms, with the aim of clarifying the cognitive and biological mechanisms that may determine whether or not clinical psychopathology develops. It is well known that long-term exposure to stress is involved in and increases the risk of mental disorders. However, the factors that influence stress levels and regulatory mechanisms of the brain are not well understood. Some of the previously suggested plausible biological and cognitive contributing factors to the stress and resilience response, such as cortisol reactivity and perceived anxiety, have been investigated before. For instance, recent findings point to a blunted cortisol stress response and greater perceived anxiety in a dose–response relationship in children * Marta Rapado-Castro mrapado@iisgm.com

D. Leander Rimmele, Theresa Schrage, Lisa Lebherz et al.

Abstract Background We aimed to identify groups of patients with similar health status after stroke, assessed by patient reported outcome measures (PROMs), to improve initial risk stratification. Methods In a prospective study, inpatients were recruited during acute stroke treatment. Demographics, history, and cardio-vascular risk factors were assessed at baseline. Self-reported functional status, physical and mental health as well as anxiety and depressive symptoms were assessed 3 and 12 months after stroke and used to identify latent classes. The association of patient characteristics with latent class membership was investigated with multinomial logistic regression. Results Of the 650 patients included with a mean age of 75 years and 48% female, 70% had ischemic, 6% hemorrhagic strokes, and 24% transient ischemic attacks. Median NIHSS on admission was 2 (IQR:0,5). Values of PROMs remained comparable at 3 and 12 months. A three-class model was developed, differentiating between patients with mildly (75%), moderately (17%), and severely (8%) impaired self-reported health status. Adjusted for univariately significant baseline characteristics, initial NIHSS distinguished mild- from moderate-, and moderate- from severe-class-membership (p < 0.001). Length of inpatient stay (p < 0.001;OR = 1.1), diabetes (p = 0.021;OR = 1.91), and atrial fibrillation (p = 0.004;OR = 2.20) predicted allocation to the moderately vs. mildly affected class. Conclusions Grading stroke patients by a standard set of PROMs up to 1 year after stroke allows to distinguish the diverse impact of baseline characteristics on differently affected groups. In addition to initial stroke severity, longer inpatient stay, presence of diabetes and atrial fibrillation correlate with greater impairment of self-reported health in the less affected groups. Trial registration http://www.ClinicalTrials.gov ; Unique identifier: NCT03795948 .

Stephanie Kandsperger, Irina Jarvers, Angelika Ecker et al.

Background: Adolescents presenting in a child and adolescent psychiatric emergency service show various psychiatric disturbances, most commonly suicidal ideation, suicide attempts, and non-suicidal self-injury (NSSI). It was postulated that especially disturbed emotion regulation contributes to self-injurious behavior of young people. This study aims to investigate the relevance of emotional reactivity (ER), as part of emotion regulation, during an acute crisis, how it relates to self-injurious behavior reinforcement and how a family as well as peers' history of self-injurious behavior are associated with self-injurious behavior of presenting adolescents. Additionally, crisis-triggering background factors were evaluated from the perspective of patients and their caregivers.Methods: A consecutive sample of 86 adolescents aged 11–18 years presenting to the emergency outpatient department due to self-injurious thoughts and behavior received a pretreatment psychiatric evaluation. Among other psychometric measures and structured clinical interviews, ER was measured via the Emotion Reactivity Scale (ERS). Family-related aspects were collected both through evaluation of history and through questionnaires filled in by custodians or parents.Results: Data analysis revealed that suicidal ideation was significantly related to family history with self-injurious behavior in comparison with a family background without such a history. A significant positive correlation was apparent between the ERS sensitivity score and occurrence of NSSI within the past year. A relationship between the ERS and distinct types of reinforcement as a motivation factor for NSSI was found. Post-hoc tests revealed a significant difference between boys and girls when no positive peers' history is present with boys having lower ERS scores than girls, but no difference when both groups had friends engaging in self-injurious behavior. There was only moderate agreement between parents and their children in naming reasons for the current crisis involving NSSI.Conclusion: Emotional regulation, especially ER, has an influence on patients' acute psychiatric symptomatology and when experiencing an acute crisis should be brought into focus early at psychiatric assessment. A history of self-injurious behavior taken from patient's family members and close circle of friends and agreement on reasons for the crisis should be routinely included in the exploration of a patient presenting with self-injurious behavior.

Arturo Tamayo, Arturo Tamayo, Timo Siepmann et al.

Posterior circulation involves the vertebrobasilar arteries, which supply oxygen and glucose to vital human brainstem structures and other areas. This complex circulatory- perfusion system is not homogenous throughout the day; rather, its hemodynamic changes rely on physiological demands, ensuring brainstem perfusion. This dynamic autoregulatory pattern maintains cerebral perfusion during blood pressure changes. Accumulative evidence suggests that activity within the autonomic nervous system is involved in the regulation of cerebral blood flow. Neither the sympathetic nor parasympathetic nervous systems work independently. Functional studies have shown a tight and complicated cross talk between these systems. In pathological processes where sympathetic stimulation is present, systemic vasoconstriction is followed, representing the most important CNS parasympathetic trigger that will promote local vasodilation. Stroke is a clear example of this process. The posterior circulation is affected in 30% of strokes, causing high morbidity and mortality outcomes. Currently, the management of ischemic stroke is focused on thrombolytic treatment and endovascular thrombectomy within an overall tight 4.5 to 6 h ischemic time window. Therefore, the autonomic nervous system could represent a potential therapeutic target to modulate reperfusion after cerebral ischemia through vasodilation, which could potentially decrease infarct size and increase the thrombolytic therapeutic ischemic window. In addition, shifting the autonomic nervous system balance toward its parasympathetic branch has shown to enhance neurogenesis and decrease local inflammation. Regretfully, the vast majority of animal models and human research on neuromodulation during brain ischemia have been focused on anterior circulation with disappointing results. In addition, the source of parasympathetic inputs in the vertebrobasilar system in humans is poorly understood, substantiating a gap and controversy in this area. Here, we reviewed current available literature regarding the parasympathetic vascular function and challenges of its stimulation in the vertebrobasilar system.

Yanan Bai, Yanan Bai, Quanliang Liu et al.

The emerging topic of privacy-preserving deep learning as a service has attracted increasing attention in recent years, which focuses on building an efficient and practical neural network prediction framework to secure client and model-holder data privately on the cloud. In such a task, the time cost of performing the secure linear layers is expensive, where matrix multiplication is the atomic operation. Most existing mix-based solutions heavily emphasized employing BGV-based homomorphic encryption schemes to secure the linear layer on the CPU platform. However, they suffer an efficiency and energy loss when dealing with a larger-scale dataset, due to the complicated encoded methods and intractable ciphertext operations. To address it, we propose cuSCNN, a secure and efficient framework to perform the privacy prediction task of a convolutional neural network (CNN), which can flexibly perform on the GPU platform. Its main idea is 2-fold: (1) To avoid the trivia and complicated homomorphic matrix computations brought by BGV-based solutions, it adopts GSW-based homomorphic matrix encryption to efficiently enable the linear layers of CNN, which is a naive method to secure matrix computation operations. (2) To improve the computation efficiency on GPU, a hybrid optimization approach based on CUDA (Compute Unified Device Architecture) has been proposed to improve the parallelism level and memory access speed when performing the matrix multiplication on GPU. Extensive experiments are conducted on industrial datasets and have shown the superior performance of the proposed cuSCNN framework in terms of runtime and power consumption compared to the other frameworks.

Tara van Viegen, Athena Akrami, Kate Bonnen et al.

Neuromatch Academy designed and ran a fully online 3-week Computational Neuroscience summer school for 1757 students with 191 teaching assistants working in virtual inverted (or flipped) classrooms and on small group projects. Fourteen languages, active community management, and low cost allowed for an unprecedented level of inclusivity and universal accessibility.

J. Buckholtz

Borderline personality disorder (BPD) is a serious mental illness characterized by volatility in mood, social attachments, and self-concept (1). A core feature of BPD is instability in interpersonal relationships. Individuals with BPD have intense and chaotic attachments to others, characterized by an often cyclical pattern of idealization followed by devaluation ( “ splitting ” ). The shift between these two extremes is often abrupt and seemingly out of proportion to the eliciting event. Splitting is often accompanied by intensely dysphoric emotional states, which in turn drive highly impulsive and typically maladaptive behaviors (e.g., self-injury, substance abuse, and reckless spending) (1). Interpersonal disturbances are responsible for much of the distress and impairment and are key targets for psychotherapeutic interventions. Yet despite the severe subjective distress, functional impairment, and economic burden imposed by interpersonal symptoms in BPD (2), their underlying cognitive and neurobiological mechanisms are only beginning to be identi fi ed. In the current issue of Biological Psychiatry: Cognitive Neuroscience and Neuroimaging , Siegel et al. (3) make a signi fi cant contribution to our understanding of these mechanisms by using an innovative computational cognition approach to understanding how patients with BPD generate and update moral inferences about other agents.