Browse Articles
Discover research articles across all indexed journals
Joint association of TyG index and LDL-C with all-cause and cardiovascular mortality among patients with cardio-renal-metabolic disease
Integrating D–S evidence theory and multiple deep learning frameworks for time series prediction of air quality
Abstract Accurate prediction of air quality time series data is helpful to identify and warn air pollution events in advance. Although the current air quality prediction models have made some progress in improving the accuracy of prediction, due to the impact of specific pollutants or complex meteorological conditions, these models still have the problems of low prediction accuracy, robustness and generalization ability in univariate prediction. In order to solve these problems, this study proposes a framework that integrates D–S evidence theory and a variety of deep learning models. The air quality data of three representative cities with climate characteristics in China are obtained and five indicators on air pollutants are collected. The preprocessed data are divided by time length to form short-term, medium-term and long-term input data, and MLP, RNN, CNN, LSTM, BI-LSTM and GRU models are established respectively. By comparing the performance indicators of the six models, three most suitable models are selected to predict the short, medium and long-term data respectively. Taking the prediction results and reliability as the three evidence bodies of the theory, a fusion model based on D–S evidence theory is established. For the three performance indicators MAE, RMSE and MAPE of the model, the best result of the fusion model increases the performance by 7.42%, 4.25% and 12.82% compared with the sub optimal architecture. This shows that integrating D–S evidence theory and a variety of deep learning algorithms provides an effective method to accurately predict the long-term air quality level in most urban areas.
Jointly exploring client drift and catastrophic forgetting in dynamic learning
Abstract Federated and Continual Learning have emerged as promising paradigms for the privacy-aware use of Deep Learning in dynamic environments by addressing spatial and temporal constraints on data availability. However, Client Drift and Catastrophic Forgetting are fundamental obstacles to ensuring robust performance. Existing work only addresses these problems separately, neglecting the fact that the root cause behind them, namely an unexpected shift in the data distribution, is connected. We propose a unified analysis framework for building a controlled test environment where we can jointly model spatial and temporal shifts, more closely emulating real dynamic settings. By generating a 3D landscape of the combined performance impact, we show that a moderate combination of both shifts can even improve the performance of the resulting model (“Generalization Bump”). We apply a simple and commonly used method from continual learning in the federated setting and observe this reoccurring phenomenon.
A computational framework to study the etiology of grandiose narcissism
Abstract Grandiose narcissism is characterized by ambivalent interaction behavior (i.e., grandiose self-presentation and rivalrous devaluation of others) and strong oscillations in self-esteem over time. In the light of emotional and social problems associated with these self-esteem regulation patterns and the increasing prevalence of narcissistic tendencies, causal and formalized models for prevention and intervention are needed. Here, we present a computational model of narcissistic self-esteem regulation implementing established, verbal theories of narcissism to identify key etiological and disorder-maintaining mechanisms. Across four studies, we show that parental praise and overvaluation lead to typical grandiose-narcissistic behavioral patterns (i.e., entitled self-presentation and rivalry) and strong self-esteem oscillations. Underlying these phenomena, we identify two maintaining mechanisms that offer targets for intervention and empirical falsification: tolerance development, characterized by an ever-increasing desire for social recognition, and a vicious cycle of rivalry, characterized by the frequent use of other-devaluing behavior and massive drops in self-esteem.
Effect of geographical origin, regional adaptation, genotype, and release year on winter hardiness of wheat and triticale accessions evaluated for six decades in trials
A robust strategy for overexpression of DNA polymerase from Thermus aquaticus using an IPTG-independent autoinduction system in a benchtop bioreactor
Abstract The DNA polymerase derived from Thermus aquaticus is the most widely utilized among various DNA polymerases, indicating its significant economic importance. Consequently, efforts to achieve a substantial yield of Taq DNA polymerase (Taq-pol) are ongoing. The expression of recombinant protein using T7-induced promoters presents challenges in cost-effectiveness, primarily due to the reliance on traditional induction method. Our study aims to enhance cost-efficiency, and scalability of our method for overproducing Taq-pol, particularly in comparison to traditional IPTG-induced techniques, which remain underreported in the current literature. To achieve those purposes, this work integrated the use of (1) a high copy number vector; (2) an optimized chemically defined medium; and (3) optimized fermentation conditions in a 5 L bioreactor. A total of 83.5 mg/L of pure Taq-pol was successfully synthesized in its active form, leading to a 9.7-fold enhancement in protein yield. This was achieved by incorporating glucose, glycerol, and lactose into a defined medium at concentrations of 0.1, 0.6, and 1%, respectively, under specific production conditions in a 5 L bioreactor: 300 rpm, 2 vvm, and 10% inoculant. The data collectively suggest that the strategy serves as a significant foundation for the future advancement of large-scale production of Taq-pol.
Exploring the causal influence of 731 immune cells on 4 different glaucoma subtypes using a two-sample mendelian randomization method
Sex-specific asymmetry in lumbar paraspinal muscles among chronic low back pain patients: correlation with pain duration and intensity
Abstract Chronic low back pain (cLBP) is a multifactorial condition, including paraspinal muscle asymmetry. Understanding the relationship between muscle asymmetry and cLBP and how this varies by sex is crucial for targeted interventions. From January 2022 to December 2023, 250 participants were enrolled: 117 without back pain (no-BP) (61 females, 56 males) and 133 with cLBP (69 females, 64 males). MRI assessed the cross-sectional area (CSA), functional CSA (FCSA), and fat infiltration (FI) of the paraspinal muscles, including the psoas major (PM), quadratus lumborum (QL), erector spinae (ES), and multifidus (MF), at all lumbar levels. Asymmetry indices were calculated for CSA (CAI), FCSA (FCAI), and FI (FIAI). Data were analyzed using multiple linear and logistic regression. Muscle asymmetries were observed in both sexes and groups, with significant differences in the CSA and FCSA of the QL in males and notable FI asymmetries in the cLBP group. In women with cLBP, higher ES CAI and FCAI at L5/S1 (p < 0.05) and lower PM FCAI at L4/5 (OR 0.869, 95% CI 1.008–1.150, p = 0.03) were observed. FIAI was significantly higher in cLBP for ES at L2/3 and PM, MF, and ES at L4/5 (p < 0.05). Men with cLBP had lower PM CAI and FCAI at L1/2 and L4/5 (p < 0.05), with varying FIAI levels. FIAI correlated with cLBP duration and intensity in both sexes (p < 0.05). Paraspinal muscle asymmetry, particularly in fat infiltration, is associated with cLBP and varies by sex. These findings support sex-specific approaches to managing cLBP.
Damage effects in GaN high electron mobility transistor power amplifier induced by high power microwave pulses
Diaphragm and sternocleidomastoid muscle activity with increasing inspiratory pressure loads in people after stroke
Author Correction: Trends in use of antiseizure medication and treatment pattern during the first trimester in the German Embryotox cohort
Aβ42 induces stress granule formation via PACT/PKR pathway
The clinical performance of robotic assisted navigation system versus conventional freehand technique for percutaneous transthoracic needle biopsy
Incidence and factors associated with agitation in patients on mechanical ventilators in Amhara Region, North-West Ethiopia: a multi-center study
The development of an artificial intelligence auto-segmentation tool for 3D volumetric analysis of vestibular schwannomas
Constructing an origin discrimination model of japonica rice in Heilongjiang Province based on confocal microscopy Raman spectroscopy technology
Synthesis and characterization of Sb/P substituted Ge–Ga–Se–Te bulk glasses
Integrated anti-electronics for positron annihilation spectroscopy
Abstract Imaging the features of a sample using Positron Annihilation Spectroscopy (PAS) is currently achieved by rastering, i.e. by scanning the sample surface with a sharply focused positron beam. However, a beam of arbitrary shape (sculpted beam) would allow the application of more versatile single-pixel imaging (SPI) techniques. I introduce the design of a microelectronic device employing a 2D array of Zener diodes as an active positron moderator, capable of sculpting positron beams with $${6}\,\upmu \hbox {m}$$ resolution. The re-emitted positrons are accelerated towards the sample through a miniaturized electrostatic lens system and reaching $${100}\,\hbox {nm}$$ resolution. The fast switch-on ( $${90}\,\hbox {ps}$$ ) and switch-off ( $${250}\,\hbox {ps}$$ ) time of the device would enable state-of-the-art positron annihilation lifetime spectroscopy (PALS) and PAS imaging with high spatial and temporal resolution.
The microRNA-6510 as a potential tumor suppressor in head and neck cancer
Abstract Head and Neck Squamous Cell Carcinoma (HNSCC) is the sixth most common cancer worldwide, with approximately 830,000 new cases and 430,000 deaths reported annually. Due to their heterogeneity, these neoplasms differ in their clinical course and response to the therapy. Therefore, it has become imperative to identify specific biological molecules that can potentially establish novel prognostic markers or targets for molecular therapy of HNSCC. MicroRNAs are a class of short, non-coding RNAs that function as post-transcriptional regulators of genes expression. They have been shown to be directly involved in oncogenesis, acting as tumor suppressors or oncogenes. Our previous study demonstrated that miRNA hsa-miR-6510-3p is significantly downregulated in tumor tissue compared to histologically normal tissue from HNSCC patients. Its significant downregulation in tumor tissue is associated with lower chances for recovery and patient’s survival. This study aimed to determine the biological role of miR-6510-3p in HNSCC pathogenesis and its impact on biological processes occurring in cancer cells such as cell cycle, cell proliferation, migration or induction of cell death. We have also examined the impact of the miR-6510-3p on expression of cancer stem cell phenotype markers as well as on sensitivity of HNSCC cells to ionizing radiation. We observed that transfection of HNSCC cells with hsa-miR-6510-3p causes the cell cycle arrest in G2/M phase and is associated with a decrease of cell proliferation, migration and colony-forming ability of cancer cells. We have also demonstrated that hsa-miR-6510-3p induces cell death, increases the sensitivity of HNSCC cells to ionizing radiation and causes a loss of the stemness properties responsible for the occurrence of metastases and relapses of the disease. These results indicated the importance of miR-6510-3p as a marker and a driver of HNSCC disease.