Search Results (215)

This paper presents a modular embedded edge-AI teaching platform built around the STM32N6 microcontroller, designed to meet demand for low-power, real-time, deployable edge intelligence in engineering education. The platform uses a heterogeneous architecture combining an ARM Cortex-M55 core with a dedicated Neural-ART NPU, enabling efficient on-device inference for both classroom projects and vision-based competition tasks. To improve stability across multi-peripheral setups, a multi-power-domain supply architecture combines switched-mode power supplies with low-noise LDO regulators, plus dual-input power switching, reverse-current and reverse-polarity protection, overcurrent limiting, and soft-start control. High-speed modular peripherals are integrated on board—MIPI-CSI camera input, RGB display output, high-speed NOR Flash, and SPI, I2C, UART, and TIMER expansion interfaces—with an 80-pin board-to-board connector for flexible extension. A four-layer PCB layout improves signal integrity for reliable high-speed operation. Deploying a custom lightweight vision algorithm (PEPoseNet), the prototype achieves an inference-only latency of 18.4 ms and 28.31 FPS/Watt energy efficiency within a sub-3 W power budget. These results confirm the platform’s reliability as an educational training platform for embedded edge-AI computing. Full article

This study assessed the cross-modal agreement between spin-exchange relaxation-free magnetoencephalography (SERF-MEG) and conventional visual evoked potentials (VEPs) in patients with optic nerve injury using comparable pattern-reversal visual stimulation paradigms. Forty-five patients with optic neuritis (ON), ischemic optic neuropathy (ION), and traumatic optic neuropathy (TON) were enrolled, and a paired-eye design was applied to assess interocular electrophysiological changes between the affected and fellow eyes. Both modalities indicated consistent impairment of visual pathway function, reflected by reduced response amplitudes and prolonged latencies. For latency measures, MEG-derived M100 showed moderate directional concordance and a significant correlation with VEP P100, whereas amplitude measures demonstrated similar directional changes but limited quantitative correlations across modalities. Compared with the MaxP2P method, the channel-averaged method displayed higher stability and cross-modal concordance. These findings show that SERF-MEG can capture clinically relevant visual pathway impairment and may complement conventional VEP by providing additional information on cortical response patterns. Overall, SERF-MEG shows potential as an objective functional biomarker for neuro-ophthalmic assessment of optic nerve injury. Full article

Background: Diabetic peripheral neuropathy (DPN) is a common complication of diabetes, characterised by sensory dysfunction and progressive loss of small nerve fibres. Corneal nerves are among the earliest fibres affected and may serve as sensitive markers for early detection and therapeutic intervention. Sitagliptin and celery seed extract have demonstrated anti-hyperglycaemic and neuroprotective properties in experimental diabetes; however, their effects on corneal nerve morphology and function remain unknown. Purpose: The present study aimed to investigate the effects of sitagliptin and celery seed extract on corneal nerve morphology, corneal sensitivity, and sensory behaviour in a high-fat diet, streptozotocin-induced rat model of type 2 diabetes. Methods: Male Sprague Dawley rats (n = 24) were used to induce type 2 diabetes by combining a high-fat diet with streptozotocin. Diabetic rats were treated with sitagliptin (30 mg/kg per day for 4 weeks) or celery seed extract (100 mg/kg per day for 4 weeks). Fasting blood glucose levels were measured throughout the monitoring period. Corneal nerve fibre parameters, including corneal nerve fibre length (CNFL), density (CNFD), and tortuosity, were assessed in βIII-tubulin-stained whole-mount corneas. Corneal sensitivity was measured with the Cochet–Bonnet esthesiometer. Sensory behaviour was evaluated using hot- and cold-water tail immersion and an acetone drop test. Results: Diabetes significantly reduced the CNFL (39.8 ± 4.7 vs. normal: 71.7 ± 7.3 mm/mm²; p < 0.05) and CNFD (7.1 ± 0.9 vs. normal: 10.6 ± 1.2%; p < 0.05) and increased corneal nerve tortuosity (5.8 ± 0.2 vs. normal: 5.1 ± 0.2; p < 0.05). Both sitagliptin and celery seed extract significantly increased the CNFL (68.8 ± 4.5 and 65.8 ± 6.5 mm/mm², respectively) compared to the untreated diabetic group. Tortuosity was significantly decreased in both the sitagliptin (4.3 ± 0.1) and celery seed extract (4.9 ± 0.1) groups compared with the untreated diabetic group. However, the CNFD showed only a modest increase after either treatment. A significant decrease in corneal sensitivity was also observed in rats following diabetes induction (5.8 ± 0.1 in normal vs. 4.5 ± 0.1 cm in diabetic rats; p < 0.05). This was rescued with sitagliptin and celery seed extract treatment (5.5 ± 0.1 and 5.3 ± 0.1 cm, respectively; p < 0.05). Compared with the non-diabetic controls, diabetic rats showed significantly shorter withdrawal latencies in both the hot- (3.9 ± 0.3 vs. 7.6 ± 1.1 s) and cold-water tests (4.3 ± 0.5 vs. 8.1 ± 0.9 s), indicating thermal sensitivity. By contrast, diabetic rats showed a significantly longer response time in the acetone drop test (8.0 ± 0.7 vs. 3.6 ± 0.4 s), indicating altered cold sensitivity (p < 0.05) for all responses. Treatment with sitagliptin and celery seed extract significantly reversed these changes, as evidenced by increased withdrawal latency in the cold-water test and decreased response time in the acetone drop test. However, this improvement was not significant in the hot-water test. Conclusions: Sitagliptin and celery seed extract restored corneal nerve architecture, corneal sensitivity, and sensory dysfunction in diabetic rats, suggesting the therapeutic potential for DPN. Full article

Background/Objectives: Herpesviridae have been increasingly investigated as possible contributors to Alzheimer’s disease (AD) because of their neurotropism, lifelong latency, and capacity to modulate inflammatory and amyloid-related pathways Methods: This structured narrative review, based on a systematic literature search, examined original studies published between December 2020 and December 2025 on the association between human herpesviridae and AD. Searches were conducted in PubMed, Web of Science, and Scopus using a PRISMA-informed screening framework. Results: The available evidence was heterogeneous across viruses and study designs. HSV-1 emerged as the most consistently implicated virus, supported by epidemiological, biomarker, neuroimaging, and experimental studies, although contradictory findings were also reported. VZV was mainly associated with AD through vascular, inflammatory, and vaccination-related evidence, with several studies suggesting lower dementia risk after zoster vaccination. CMV, EBV, and HHV-6 showed biologically plausible but less consistent associations, whereas HHV-7 and HHV-8 were supported only by limited or indirect evidence. Across studies, the proposed mechanisms included chronic neuroinflammation, vascular injury, amyloid and tau dysregulation, host immune responses, and cumulative infectious burden. Conclusions: Overall, the literature suggests that the relationship between herpesviridae and AD is not uniform and that HSV-1 appears to be the most relevant candidate. However, methodological heterogeneity, possible reverse causation, and unequal study intensity across viruses limit firm conclusions. More robust longitudinal and subtype-specific studies are needed to clarify causal relevance. Full article

Epidermolysis bullosa (EB) comprises a group of rare inherited genodermatoses characterized by fragility and blistering of the skin and mucous membranes, chronic wounding, and significant morbidity including increased risk of squamous cell carcinoma in severe subtypes. Key unmet priorities include reducing diagnostic latency, establishing objective wound monitoring, enabling early detection of malignant transformation within chronic ulcerations, and developing therapies that durably modify disease progression. Artificial intelligence (AI) encompassing machine learning (ML), and deep learning (DL) is increasingly integrated into EB research and clinical practice to address these unmet needs. This structured narrative review synthesises current evidence on AI applications in EB spanning genetic diagnostics, wound assessment, inflammatory endotyping, drug repurposing, and emerging therapeutic technologies, and integrates evidence from registered clinical trials. In genomics, DL-based splicing prediction models and variant prioritisation frameworks accelerate pathogenic variant detection and reduce diagnostic latency. In wound care, convolutional neural networks-based platforms enable automated lesion segmentation and remote monitoring, while multimodal AI models predict healing trajectories and support stratification of wounds by chronicity. Computational transcriptomic analyses have identified candidate repurposing agents by reversing pathogenic gene expression signatures in EB tissue. Emerging convergence of AI with biosensors-integrated wound dressings and three-dimensional bioprinting of genetically corrected skin substitutes represents a transformative future direction. Translational barriers include limited EB-specific training datasets, algorithmic bias across diverse skin phototypes, the interpretability deficit of DL systems, and evolving regulatory frameworks for AI as a medical device. Expansion of internationally interoperable EB disease registries with standardised wound imaging protocols is identified as the single most impactful intervention to accelerate AI adoption. A minimum endpoint set for AI-assisted EB wound assessment, incorporating wound area trajectory, wound type classification, tissue composition, and paired patient-reported pain and itch scores, is proposed to standardise outcome reporting across future studies. Full article

Background: The glucocorticoid receptor NR3C1 exhibits antiepileptic properties, but the mechanisms governing its stability during epileptogenesis remain elusive. This study investigated whether the E3 ubiquitin ligase TRIM8 regulates neuronal hyperexcitability and epileptic activity by modulating NR3C1. Methods: We established an in vivo epilepsy model via intrahippocampal kainic acid (KA) injection and an in vitro epileptiform model using Mg²⁺-free artificial cerebrospinal fluid in primary hippocampal neurons. The roles of TRIM8 and NR3C1 were assessed using in vivo and in vitro gain- and loss-of-function approaches, alongside co-immunoprecipitation, Western blotting, immunofluorescence and whole-cell patch-clamp recording. Results: TRIM8 is significantly upregulated in hippocampal and temporal lobe neurons in epileptic mice. TRIM8 was markedly upregulated in the hippocampal neurons of epileptic mice, inversely correlating with NR3C1 levels. Mechanistically, TRIM8 interacted with NR3C1, promoting its polyubiquitination and proteasomal degradation. This TRIM8-mediated NR3C1 reduction enhanced the phosphorylation of AMPA receptor (AMPAR) subunits GluR1 (Ser831) and GluR2 (Ser880) without affecting total receptor expression. In vitro, TRIM8 overexpression exacerbated calcium dysregulation, neuronal injury, and AMPAR phosphorylation; crucially, concurrent NR3C1 overexpression rescued these effects. In vivo, knockdown of TRIM8 significantly reduced seizure frequency, prolonged the latency to the first Stage III seizure, shortened average seizure duration, and decreased total seizure burden in KA-induced epileptic mice. Electrophysiologically, TRIM8 overexpression significantly increased the frequency of spontaneous action potentials and amplitudes of spontaneous excitatory postsynaptic currents under Mg²⁺-free conditions. Furthermore, in vivo knockdown of TRIM8 attenuated KA-induced seizure severity, restored NR3C1 protein stability, and suppressed aberrant AMPAR phosphorylation in the hippocampus. Triple immunofluorescence staining showed that KA-induced epilepsy increased TRIM8 but decreased NR3C1 immunoreactivity in NeuN⁺ hippocampal neurons, and TRIM8 knockdown reversed these changes. Conclusions: TRIM8 acts as a critical driver of epileptiform activity by targeting NR3C1 for degradation, thereby disinhibiting AMPAR phosphorylation and enhancing network hyperexcitability. The TRIM8-NR3C1-AMPAR axis emerges as a previously unrecognized molecular pathway in epileptogenesis, highlighting its potential as a promising therapeutic target for epilepsy. Full article

In UAV-based remote sensing, accurate and efficient image mosaicing is crucial for achieving real-time monitoring. Traditional cloud-centric processing paradigms, however, face core scientific challenges such as high latency, bandwidth bottlenecks, and limited autonomy, making them inadequate for dynamic, real-time scenarios. To address these issues, this paper proposes an edge-computing-enabled UAV image mosaicing system. The system consists of a UAV remote sensing platform and an edge computing terminal, with the core being our novel B-SIFT-ILS algorithm. The algorithm first uses geographic coordinates for unified registration, constructs a Gaussian scale space for multi-resolution representation, and then precisely locates extrema in the Difference of Gaussian (DoG) space using a 3D quadratic function. A BANSAC algorithm is subsequently employed to refine feature points and extract stable SIFT features, and finally, Iterative Least Squares (ILS) are used to achieve seamless mosaicing. Experimental results demonstrate that, compared with classical RANSAC, the proposed method achieves superior feature sampling accuracy (rotation: 0.879, translation: 0.877) and lower latency. The ILS-based smoothing stage effectively eliminates noise and ghosting without introducing gradient reversal, performing comparably to deep learning methods while significantly outperforming direct averaging and Gaussian approaches. On the NVIDIA Jetson Orin NX edge terminal, a single processing instance requires only 1124 ms, highlighting its strong potential for real-time, low-latency, and autonomous mosaicing tasks. Future research will focus on extending the approach to non-planar terrains and implementing adaptive parameter tuning for the BANSAC algorithm. Full article

This study aimed to characterize cortical dysfunction and frequency-specific network reorganization following optic nerve injury using spin-exchange relaxation-free magnetoencephalography (SERF-MEG), and to assess the potential of MEG-derived multiscale features as sensitive functional biomarkers for clinical evaluation. In this prospective case–control study, SERF-MEG recordings were acquired during a pattern-reversal visual stimulation paradigm. Time-domain evoked components (M100/M135), global electrophysiological indices, energy-based metrics, and alpha- and beta-band phase-based functional connectivity were extracted. Network topology was quantified using graph-theoretical measures, including global and local efficiency, clustering coefficient, and assortativity. Group-level differences between patients and healthy controls were statistically analyzed. Patients showed significantly reduced M100/M135 amplitudes, prolonged M100 latency, and a lower early-component energy ratio. Functional connectivity was significantly decreased in the alpha and beta bands, accompanied by reduced global and local efficiency, mean strength, and clustering coefficient. Seed-based analyses revealed reduced connectivity predominantly in occipito-parietal and occipito-temporal pathways. SERF-MEG provides sensitive identification of cortical- and network-level functional impairments following optic nerve damage. MEG has significant clinical potential for disease diagnosis and therapy monitoring, providing a novel objective assessment tool for neuro-ophthalmological disorders. Full article

Latently infected resting CD4+ T cells represent the principal barrier to HIV eradication. Their immunological quiescence renders them invisible to combination antiretroviral therapy (cART) and host immune surveillance, sustaining a reservoir that mandates lifelong treatment. The ‘shock and kill’ strategy seeks to reverse this latency using pharmacological latency-reversing agents (LRAs) so that immune effectors and cART can eliminate reactivated cells. In this narrative, structured review, we examine histone deacetylase inhibitors (HDACis) as LRAs from a medicinal chemistry perspective. The review places particular emphasis on structure–activity relationship (SAR), isoform selectivity, and the mechanistic basis of differential clinical performance, synthesizing evidence from preclinical, ex vivo, and clinical studies published between 2010 and 2026. Four structural classes of HDACis—hydroxamic acids, benzamides, cyclic depsipeptides, and short-chain fatty acids—differ substantially in isoform selectivity, potency, pharmacokinetics, and tolerability. No single agent has achieved statistically significant reservoir reduction in clinical trials, highlighting the apparent inadequacy of the ‘shock’ phase alone and suggesting a need for complementary ‘kill’ strategies. Rational design of HDACis informed by isoform-selective SAR, combined with emerging combination LRA strategies and immunological ‘kill’ components, represents a promising direction toward a functional HIV cure, though substantial translational hurdles remain. Full article

Background/Objectives: AIDS is a serious threat to human health and remains incurable; however, EK-16A, an ingenol derivative, shows promise as a functional cure. In this study, we aimed to extract EK-16A from Euphorbia kansui, used in traditional Chinese medicine, to develop an oral self-nanoemulsifying drug delivery system (SNEDDS) for EK-16A and evaluate it in vivo. Methods: EK-16A was purified by SFC combined with conventional extraction. The optimal SNEDDS formulation was selected by emulsification and stability testing. Pharmacokinetics, metabolomics, and proteomics were used for in vivo evaluation. Results: 1. The extraction yield of EK-16A was four times higher than that of the conventional process; the extraction scale was increased by 25 times, and the purity of EK-16A reached 98.0%. 2. EK-16A is a BCS Class IV compound with low solubility and permeability. The compound’s content degraded to 49.8% after 3 months at 25 °C/60% RH. The EK-16A SNEDDS formulation A#1 showed no degradation after 3 months at 40 °C/75% RH. The absolute bioavailability after oral administration of formulation A#1 in rats was 0.445%. 3. The proteomics results showed that EK-16A significantly downregulated the PI3K-AKT signaling pathway in SHIV-infected rhesus macaques. Specifically, all 11 identified differential proteins were significantly downregulated. Conclusions: 1. The extraction process for EK-16A features high yield, purity and large scale. 2. The SNEDDS formulation enhances the stability of EK-16A and successfully delivers this low-solubility and permeability compound into the systemic circulation. 3. Proteomics analysis revealed that EK-16A significantly downregulates the PI3K-AKT signaling pathway in SHIV-infected rhesus macaques. However, further experiments, such as measuring plasma viremia and cell-associated SHIV RNA, are needed to confirm this mechanism. Full article

Antiretroviral therapy (ART) effectively suppresses HIV-1 replication but does not purge the latent HIV-1 reservoir. Strategies aimed at HIV-1 latency reversal and subsequent elimination of infected cells are being explored. Targeting the inhibitor of apoptosis proteins (IAP) and DEAD-box polypeptide 3 (DDX3) RNA helicase reduces the HIV-1 reservoir ex vivo. However, the mechanisms driving apoptosis of HIV-1 infected cells remain unclear. Here, we uncovered the mechanism regarding HIV-1 transcriptional activation and induction of apoptosis specific for HIV-1 infected cells using an acute in vitro infection model. Inhibition of IAP by second mitochondrial-derived activator of caspases mimetic (SMACm; AZD5582) resulted in activation of non-canonical NF-κB pathway (RelB/p52) that induced HIV-1 transcription, confirming previous reports, whereas inhibition of DDX3 sensitized HIV-1 infected cells for apoptosis (DDX3i; FH1321). Transcriptome analysis revealed that HIV-1 actively suppressed apoptosis-related genes in HIV-1 infected cells. SMACm treatment resulted in a broad induction of these genes irrespective of infection. Notably, DDX3 inhibition specifically restored the expression of the majority of HIV-1 suppressed genes, and when combined with SMACm, restored almost all HIV-1 downregulated genes, thereby rendering HIV-1 infected cells sensitive to apoptosis. Thus, our data strongly suggest that inhibition of host factors IAP and DDX3 not only induces activation of HIV-1 transcription but also restores HIV-1 suppressed apoptotic processes in infected cells. Full article

Cognitive impairment induced by a high-fat high-fructose diet (HFFD) is associated with gut–liver–brain axis dysfunction, yet whether polysaccharide intervention can modulate this axis to achieve cognitive rescue remains unexplored. This study investigated whether Elaeagnus angustifolia polysaccharide (EAP) is associated with protection against HFFD-induced cognitive decline by modulating this axis. Male C57BL/6J mice (n = 15/group) received Control, HFFD, HFFD + LEAP (300 mg/kg/day EAP), or HFFD + HEAP (800 mg/kg/day EAP) for 14 weeks. HEAP improved spatial memory, reducing escape latency by 31.2% on day 5 (p < 0.01). Multi-omics and histopathological analyses revealed that EAP was dose-dependently associated with restructuring of the gut microbiota, expanding Muribaculaceae and other SCFA-producers while suppressing pathobionts, thereby reversing the Firmicutes/Bacteroidota ratio from 1.71 to 0.94 (p < 0.01). Elevated cecal, hepatic, and cerebral acetate, propionate, and butyrate (p < 0.01) were associated with improved intestinal barrier integrity, attenuated systemic LPS translocation, and reduced hepatic inflammation and changes consistent with normalization toward control levels of PPARα/γ signaling. These peripheral improvements were accompanied by changes in the hippocampus, where EAP suppressed IBA-1 microglial activation (from 4.5-fold to 2.1-fold of control, p < 0.01) and IL-6/TNF-α signaling, changes in neurotransmitter balance (Glu, 5-HT, DA), and preserved postsynaptic density ultrastructure and PSD-95 expression (p < 0.01). These findings support a role for EAP in modulating the gut–liver–brain axis and may help prevent diet-related cognitive impairment, supporting its development as a microbiome-targeted functional food ingredient. Full article

Large-scale shell-and-tube heat exchangers operate for extended periods, critically affecting the energy efficiency and safety of hydrogen production processes. However, online condition monitoring on industrial distributed control systems (DCSs) is often hindered by an engineering trilemma: high-fidelity mechanistic models incur prohibitive computational latency; static constant-parameter models suffer from severe systematic bias; and purely data-driven models risk yielding non-physical predictions under out-of-distribution scenarios such as variable-load operations. To address these challenges, this study proposes a physics-guided adaptive digital twin tailored to high-noise industrial DCS environments. Energy conservation and the counterflow logarithmic mean temperature difference (LMTD) relation are embedded as hard constraints in a lightweight reduced-order model (ROM). On this basis, a closed-loop online adaptation strategy—comprising physical-bound checking, window-wise inverse estimation, anomaly rollback, and exponentially weighted moving average (EWMA) smoothing—treats the overall heat transfer coefficient U as an equivalent time-varying parameter that co-evolves with operating regimes. Validation on real plant DCS data under variable-load conditions shows that, compared with a conservative fixed-U baseline, the proposed online update eliminates massive systematic overestimations (up to tens of degrees Celsius) and suppresses inversion oscillations caused by small cold-side temperature differences and sensor noise. Relative to an overfitting-prone data-driven baseline, the framework retains millisecond-level inference latency while enforcing thermodynamic feasibility, thereby establishing a dynamic healthy baseline. This baseline provides a proxy indicator for distinguishing load-induced reversible variations from potential degradation-related residual trends. Full article

Depression is a complex mental and neurological disorder and has become one of the most serious public health issues in modern society. Trihexyphenidyl (THY) is a traditional drug used to treat Parkinson’s disease. Recent studies have suggested that it may play a role in regulating neurotransmitters and protecting neurons, but its potential for treating depression has not been fully explored, and how it works remains unclear. Therefore, we examined the effects of THY on depression-like behaviors in zebrafish caused by chronic unpredictable stress (CUS). Our results showed that THY significantly attenuated the CUS-induced decrease in exploratory behavior and shortened the CUS-induced increase in latency time. At the tissue level, THY effectively attenuated the thinning of the optic tectum and the loss of Nissl bodies caused by CUS. In addition, THY reversed the CUS-induced increase in stress hormone levels and reduction in neurotransmitter content. Through network pharmacology and transcriptome sequencing analysis, we found that the mechanisms underlying depression-like behaviors and the antidepressant effects of THY might be related to the MAPK signaling pathway. Further experiments showed that THY regulated the CUS-induced activation of the MAPK signaling pathway, improved the abnormal activation of microglia and damage to astrocytes, and reduced the expression of pro-inflammatory factors, thereby easing neuroinflammation and improving depression-like behaviors. In summary, this study explored the potential mechanism of THY ameliorating depressive-like behaviors and provided basic theoretical evidence. Full article

This study presents a systematic empirical evaluation of monolithic and distributed database architectures deployed in Kubernetes environments. As containerized and cloud-native infrastructures become increasingly prevalent, understanding the performance implications of running stateful data systems under orchestration platforms has become critical. We evaluate five widely used database systems—PostgreSQL, MySQL, MongoDB, Redis, and Cassandra—using standardized workload generation frameworks, including pgbench, sysbench, YCSB, redis-benchmark, and cassandra-stress. Controlled experiments were conducted across varying concurrency levels and workload types to measure throughput, latency, and scalability in both single-node and distributed deployments. Redis achieves a maximum throughput of 4.2 million operations per second with sub-millisecond latency. In contrast, Cassandra delivers 214,743 distributed read operations per second at ONE consistency, approaching Redis’s non-pipelined baseline throughput (257,732–262,467 ops/sec) within a Kubernetes cluster. The write throughput of Cassandra decreases by 45.2% when the consistency level is elevated to QUORUM, accompanied by an elevenfold increase in run-to-run variability (CV from 7.1% to 84.7%), indicating that the consistency level is the primary performance determinant in distributed systems. PostgreSQL experiences a 72% decrease in write throughput in Kubernetes (74,072 → 20,805 TPS). In contrast, MySQL PXC anomalously attains a 37.3% increase in write throughput in Kubernetes compared to its monolithic deployment—the sole reversal noted among the five systems. These findings underscore a critical trade-off between vertical efficiency and horizontal scalability, illustrating that hybrid database architecture can be an effective solution for contemporary cloud-native applications compared to either paradigm independently. Full article