Browse Articles

Goal-guided greedy experience replay-enhanced reinforcement learning for efficient autonomous navigation

Proto-memformer: deformable memory transformer for Parkinson’s MRI classification

Abstract This paper proposes a Prototype-Guided Deformable Memory Transformer (Proto-MemFormer) model for Parkinson’s Disease (PD) MRI classification. In the encoding stage, the model integrates a prototype-guided memory mechanism with a deformable attention structure to dynamically aggregate local morphological features and global semantic information. In the decoding stage, a position-calibrated retrieval module is introduced to enhance cross-sample feature alignment and discriminative representation. Experiments conducted on two public datasets, NTUA-Parkinson and PPMI, demonstrate that the proposed model achieves Accuracy, Precision, Recall, F1-Score, and AUC of 93.45%, 93.72%, 93.21%, 93.46%, and 94.81%, respectively, on the NTUA-Parkinson dataset, outperforming current state-of-the-art deep learning methods. Moreover, in the hyperparameter and training set scaling experiments, the model exhibits performance fluctuations of less than 3%, verifying its stability and robustness under different data and environmental conditions.

Cystathionine β-synthase is inhibited by epinephrine and norepinephrine over-secretion via NF-κB activation in stress-induced hyperhomocysteinemia

Abstract Stress can elevate plasma homocysteine (Hcy) levels, a key factor in the development of various diseases, leading to hyperhomocysteinemia (HHcy). The enzyme Cystathionine β-synthase (CBS) functions in the trans-sulfuration pathway, which links the methionine cycle to glutathione synthesis. This pathway is essential for the metabolism of homocysteine. However, the precise mechanism by which stress regulates hepatic CBS expression remains unclear. The present study aimed to elucidate the molecular mechanism by which CBS expression is regulated in the livers of restraint-stressed rats. Our results showed that stress-induced over-secretion of epinephrine and norepinephrine (E/NE) activated the β-adrenergic receptor, leading to elevated serum IL-6 levels. Treatment of rat primary hepatocytes with IL-6 for 1 h suppressed both CBS activity and Cbs mRNA expression, concomitant with an enhancement of Sp3-mediated transcriptional repression. Furthermore, we found that IL-6 activates NF-κB via the tyrosine phosphorylation pathway, which in turn potentiates Sp3-mediated repression of Cbs transcription. These findings suggest that E/NE contributes to stress-induced HHcy by inhibiting Cbs transcription through the upregulation of Sp3, and that the IL-6/NF-κB axis plays a critical role in the dysregulation of Hcy metabolism.

An explainable multimodal temporal deep learning framework for intrusion detection (EMT-IDNet) in IoT environments

Population-based comparison of post-acute sequelae of COVID-19 and health-related quality of life across pandemic periods: Omicron era versus early pandemic

Abstract Post-COVID-19 syndrome (PCS) may vary across pandemic phases. We compared the prevalence of post-acute COVID-19 symptom clusters following SARS-CoV-2 infection in the early pandemic and during the Omicron era using a population-based approach. The EPILOC study enrolled individuals aged 18–65 years who tested PCR-positive for SARS-CoV-2 during the early pandemic, between October 2020 and April 2021, in defined geographic regions within Baden-Württemberg. EPILOC-Omicron used an identical design for individuals infected between June and July 2022. Participants completed a standardised questionnaire assessing sociodemographics, lifestyle, symptoms, and health-related quality of life (SF-12). PCS was defined as ≤ 80% recovery of general health or work capacity plus at least one new moderate-to-strong symptom, both compared to before index infection. Generalised linear models adjusted for age, sex, and education were used to estimate relative risks (RRs). We analysed data from 11,710 EPILOC (24% response) and 12,560 EPILOC-Omicron participants (17% response). Participants were similar in age and sex distribution. Vaccination before infection differed markedly (< 3 vs. > 92%). PCS prevalence was 29.6% in EPILOC and 14.5% in EPILOC-Omicron. Predictors of PCS were similar for both. Symptom clusters were consistently less frequent after infection during the Omicron era (e.g., fatigue RR = 0.54; neurocognitive impairment RR = 0.53; chest symptoms RR = 0.47; smell/taste disorder RR = 0.17). Health-related quality of life was similar in PCS cases of both cohorts (mean SF-12 Physical 40.3 vs. 41.0, mean SF-12 Mental 38.9 vs. 40.7). PCS was less common following SARS-CoV-2 infection in the Omicron era compared to the early pandemic period. However, affected individuals continue to experience substantial, comparable impairment in hrQoL across both pandemic periods.

Does agricultural green finance help reduce agricultural carbon emission intensity: an empirical analysis based on 30 provinces in China

Inhibition of the Atypical Kinase WNK1 as a Therapeutic Strategy in TAL-related T-cell Acute Lymphoblastic Leukemia

Driver mutations in T-cell acute leukemia (T-ALL) rarely affect druggable kinases. However, these kinases can be aberrantly activated or repressed as secondary oncogenic events. Thus, integrating unbiased phosphoproteomics with genomic approaches may offer novel opportunities for target discovery and therapeutic interventions. In our study, we identified WNK1 as a potential target in T-ALL by pairing a list of vulnerable kinases with data from a phosphoproteomic screen of T-ALL cell lines. We subsequently validated WNK1 by loss-of-function-based studies and tested WNK inhibitors in several in vitro and in vivo T-ALL models and clinical T-ALL samples. We showed that therapeutic WNK1 repression promotes polyploidy, resulting in cell proliferation arrest, and morphometric changes, such as incomplete cell division or chromosome segregation through altered mitotic spindle assembly and abscission defects. Furthermore, we found that WNK1 is overexpressed in the TAL1/2-related subgroup, but not in normal thymus or lymph nodes, suggesting a potential translational area for clinical exploitation in poor-prognosis T-ALL carrying PTEN mutations and del(6q). Our work also reports a functional contribution of WNK1 in the leukemia establishment and progression. Structurally WNK1 is an atypical serine/threonine kinase that diverge from canonical kinases by lacking the conserved lysine in subdomain II, instead featuring a cysteine in subdomain I, which is critical for ATP-binding. This unusual structural configuration creates a distinct ATP-binding pocket with limited sequence similarity to conventional kinases, offering a unique opportunity to develop highly selective small-molecules. Targeting this atypical ATP-domain could thus provide a therapeutic advantage and broaden the treatment landscape for T-ALL.

Correction: The patterns of acetylcholinesterases during developmental stages of Aedes aegypti and their susceptibility toward insecticides in egg stage

TGFβ-PDL1 signaling in neutrophils preserves lung barrier during hyperinflammation

The hyperinflammatory syndromes in critically ill patients, including trauma, sepsis, and acute lung injury, are characterized by dysregulated neutrophil responses that contribute to tissue damage and poor outcomes. Using murine models of cytokine storm induced by trauma and lung injury, we identified transforming growth factor β (TGFβ) as a central regulator of immune checkpoint in neutrophils. TGFβ signaling modulates neutrophil activation and upregulates the expression of programmed death-ligand 1 (PDL1). Disruption of TGFβ signaling during hyperinflammation restores the migratory capacity of neutrophils but leads to excessive activation, severe pulmonary tissue damage, and increased susceptibility to spontaneous bacterial infection in the lung. Mechanistically, PDL1 expression alters neutrophil behavior within lung capillaries, promoting intravascular clustering and restricting tissue infiltration. Targeted deletion of PDL1 in neutrophils reverses hyperinflammation-induced clustering, restores effective trafficking to infectious foci, and enhances host-protective immune function while limiting pathological neutrophil hyperactivation. These findings define a TGFβ-PDL1 regulatory axis that restrains the pathogenicity of neutrophils during hyperinflammation, revealing a checkpoint mechanism that balances host defense and tissue integrity.

Correction: Accumulation of human full-length tau induces degradation of nicotinic acetylcholine receptor α4 via activating calpain-2

Quantum machine learning for predicting anastomotic leak: a clinical study

Abstract Anastomotic leak is a life-threatening complication following colorectal surgery. This study benchmarks Quantum Neural Networks (QNNs) against hyperparameter-tuned classical models (logistic regression, multi-layer perceptrons, boosting algorithms) for anastomotic leak prediction. Using a 200-patient clinical dataset strictly bounded by a priori medical constraints, we simulated QNNs with ZZFeatureMap encoding and EfficientSU2/RealAmplitudes ansatze under realistic hardware noise. To ensure statistical reliability, performance metrics were averaged across 10 independent optimization runs. The EfficientSU2-BFGS configuration achieved the highest mean AUC of $$0.797 \pm 0.024$$ , while RealAmplitudes with CMA-ES maximized Average Precision ( $$0.504 \pm 0.121$$ ). Crucially, at a fixed, clinically necessary sensitivity of $$83\%$$ , specific QNN configurations achieved significantly higher specificity (up to $$66\%$$ ) and Negative Predictive Value (up to $$96\%$$ ) compared to classical models (maximum $$44\%$$ and $$94\%$$ , respectively), effectively minimizing false positives. However, classical models maintained superior probability calibration for continuous risk stratification. We conclude that QNNs offer robust discriminative performance for clinical screening, warranting further validation on larger, independent cohorts.

Evaluating the effectiveness and roadside noise of alternative transverse rumble strip designs

Abstract This study conducted field measurements to evaluate the performance of five promising traditional and alternative transverse rumble strip designs in generating effective in-vehicle noise level increases to alert distracted and drowsy drivers while reducing roadside noise levels. The five transverse rumble strip designs consisted of one baseline traditional design and four promising designs including shorter panel, angled, staggered, and sinusoidal designs. The in-vehicle and external noise measurements were conducted following the SAE J1477 standard and the AASHTO Statistical Isolated Pass-by (SIP) method, respectively. The noise measurements were collected using thirteen different vehicles with variations in type, make, size, weight, engine, and model including gasoline, hybrid, and electric sedans as well as SUVs, minivan, pick-up, box, and heavy semi-trailer trucks to represent all vehicles on U.S. roads. The findings of this study confirmed that all five transverse rumble strip designs generated average in-vehicle noise level increases between 3 and 15 dBA that fulfill the NCHRP recommendations for alerting distracted drivers. The analysis results also indicated that the four promising transverse rumble strip designs reduced roadside noise levels by a range of 18% (1.8 dBA) to 77% (7.7 dBA) compared to the baseline traditional design. These research findings provide state Departments of Transportation (DOTs) with new effective designs that maintain the roadway safety benefits of transverse rumble strips while decreasing their roadside noise levels and related complaints from nearby residents.

Coupling mechanism of dynamic incidence angle and multipulse accumulation in femtosecond laser ablation of complex spatial surfaces

Resonance frequency versus fixed 0.1 Hz breathing in HRV biofeedback: a four-week randomized comparison

Abstract Heart rate variability biofeedback (HRVB) is increasingly examined as an adjunctive method for reducing psychological symptoms. Standard protocols involve identifying each participant’s resonance frequency (RF); however, it remains unclear whether this individualized calibration offers advantages over training at a fixed breathing rate of 0.1 Hz. We conducted a randomized trial with three groups (RF, 0.1 Hz, Control) to compare the effectiveness of a four-week HRVB intervention using an individually determined RF versus a fixed breathing frequency of 0.1 Hz. Participants ( N  = 88) completed pre- and post-intervention assessments of perceived stress, anxiety, depressive symptoms (DASS-21), and resting HRV. Both HRVB groups showed significant reductions in stress, anxiety, and depressive symptoms relative to the control group, with no meaningful differences between them. Bayesian analysis provided anecdotal evidence favoring the null hypothesis of no difference between groups. The intervention did not produce significant changes in resting HRV. The present study does not permit firm conclusions regarding whether breathing at the RF confers additional benefits over breathing at a fixed rate of 0.1 Hz or whether the two approaches differ in effectiveness. Future studies aimed at detecting small differences between these approaches should include larger samples. Trial registration : The study was retrospectively registered at isrctn.com (Identifier ISRCTN17808563 date of first registration 20/01/2026, https//doi.org/10.1186/ISRCTN17808563). The full trial protocol and statistical analysis plan can be accessed via the registry.

Hydrogen storage efficiency and optoelectronic properties of structutal, mechanical, and dynamical stable perovskite (Li/Na/K)BH3 hydrides for sustainable energy applications

Effects of player number and pitch size on internal and external load in small-sided Ultimate Frisbee games

Concentrations of heavy metals and biogenic elements in needles of Norway spruce and Scots pine from Polish peatlands

Abstract The study characterised the content and spatial variability of heavy metals and selected biogenic elements (Cd, Co, Cr, Cu, Ni, Pb, Zn, Al, Ca, Fe, K, Mg, Mn, Na, P, N, C) in one-year-old needles of Norway spruce and Scots pine from 262 peatlands in Poland in the light of previous studies and norms, and to determine the relationship between the content of elements in needles and the types of Histosol (fibric, hemic, sapric). The spatial variation in heavy metal content in needles is not related to southern regions of Poland with high human footprint index which suggests that episodic local sources of pollution and transboundary transport of air pollutants are likely to influence their bioaccumulation. It is also not related to the type of Histosol or the pH of water and soil, which indicates that heavy metals accumulate mainly passively from the atmosphere. Despite exceeding the optimal values for some elements, heavy metal levels do not pose a toxicological risk. Elements such as chromium in spruce and cadmium in pine can serve as markers of the impact of industrial emissions on peatland ecosystems. The concentrations of the most studied elements in Poland were lower than in other European countries.

A novel lexicon dictionary and CNN-LSTM employed hybrid approach for sentiment detection of COVID-19 vaccines

Determinants of scabies among households in Fagita Lekoma district, Northwest Ethiopia: a mixed–method study

High-depth whole genome sequencing of premalignant breast lesions reveals rearrangement hotspots and personalized management opportunities

Abstract Ductal carcinoma in situ is a non-obligate precursor lesion of breast cancer. Often detected by mammography, most cases are managed through surgical and/or radiotherapy approaches. Today, it is not possible to predict which patients will progress to invasive disease. Here, we evaluate high-depth whole-genome sequenced ductal carcinoma in situ, enriched for high-grade clinical lesions, to understand whether deep WGS could reveal biological insights and/or personalized therapeutic vulnerabilities that may be targetable. We find genomic locations that are likely susceptible to producing the initiating lesion for structural variations prone to subsequent evolution, termed SHOREs. We additionally highlight individualized therapeutic potential that would otherwise not be appreciable without whole genome sequencing. We posit that holistic whole genome sequencing profiling could offer a more precise stratification approach, discerning higher-risk cases for prospective clinical studies on personalized therapies, from truly low-risk cases suitable for active monitoring.