Search Results (44)

Diabetes is a developing global health concern that cannot be cured, necessitating frequent blood glucose monitoring and dietary management. Photoacoustic Spectroscopy (PAS) in the mid-infrared (MIR) region has recently emerged as a viable noninvasive blood glucose monitoring technique. However, MIR-PAS confronts significant challenges: (i) Water absorption, which reduces light penetration, and (ii) interference from other blood components. This paper systematically analyzes the background of photoacoustic signal generation and proposes a differential PAS (DPAS) in the MIR region for removing the background signals arising from water and other interfering components of blood, which improves the overall detection sensitivity. A detailed mathematical model with an explanation for choosing two suitable MIR quantum cascade lasers for this differential scheme is presented here. For single-wavelength PAS (SPAS), a detection sensitivity of $1.537$ µPa mg⁻¹ dL was obtained from the proposed model. Alternatively, $2.333$µPa mg⁻¹ dL detection sensitivity was found by implementing the DPAS scheme, which is about 1.5 times higher than SPAS. Moreover, DPAS facilitates an additional parameter, a differential phase shift between two laser responses, that has an effective correlation with the glucose concentration variation. Thus, MIR-based DPAS could be an effective way of monitoring blood glucose levels noninvasively in the near future. Full article

Natural Language Processing (NLP) is a valuable technology and business topic as it helps turn data into useful information with the spread of digital information. Nevertheless, there are some difficulties in its use, including the language’s complexity and the data quality. To address these challenges, in this study, the researchers first performed a series of ablation experiments on 14 models derived from various variations in Deep Learning (DL) methods, including A Lite BERT (ALBERT) together with Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Max Pooling layer, and attention mechanism. Subsequently, they proposed an ALBERT-cascaded CNN hybrid model as an effective method to overcome the related challenges by evaluating the performance results obtained from these models. In the proposed model, a transformer architecture with parallel processing capability for both word and subword tokenization is used in addition to creating contextualized word embeddings. Local and global feature extraction was also performed using two 1-D CNN blocks before classification to improve the model performance. The model was optimized using an advanced hyperparameter optimization tool called OPTUNA. The findings of the experiment conducted with the proposed model were obtained based on Amazon Fashion 2023 data under 5-fold cross-validation conditions. The experimental results demonstrate that the proposed hybrid model exhibits good performance with average scores of 0.9308 (accuracy), 0.9296 (F1 score), 0.9412 (precision), 0.9182 (recall), and 0.9797 (AUC) in the validation dataset, and scores of 0.9313, 0.9305, 0.9414, 0.9199, and 0.9800 in the test dataset. In addition, comparisons of the model with models in studies using similar datasets support the experimental results and reveal that it can be used as a competitive approach for solving the problems encountered in the NLP field. Full article

Radar backscatter from large-scale scenarios plays a crucial role in remote sensing applications. However, due to the diversity and heterogeneity of the natural environment, traditional empirical methods which rely on simplified physics and a limited set of parameters, fail to adequately model land backscatter, thus exhibiting significant limitations. While purely data-driven deep learning (DL) methods offer flexibility, they often struggle to ensure physical consistency and effectively generalize to unseen scenarios. To address these issues, we propose a novel knowledge-aided (KA) DL-based method (called KADL) in this paper for predicting the radar backscatter from large-scale scenarios. The proposed KADL is implemented in three parts. First, based on multi-source remote sensing data, the dielectric properties of land surface, i.e., soil moisture and leaf area index (LAI) are incorporated as priori physical knowledge into the Multi-Feature Clutter Dataset (MFCD) to obtain initialized input. Second, a knowledge perception module (KPM) is introduced into the cascaded deep neural network (DNN) solver to exploit the representative features within the inputs. Third, an efficient knowledge-weighted fusion (KWF) strategy is developed to further enhance the discriminative features and simultaneously suppress the non-informative features. For better comparison, we refitted the specific empirical models based on the measured data and introduced an advanced nonhomogeneous terrain clutter model (termed ANTCM) derived from our previous work. Extensive experiments conducted on the measured data demonstrate that KADL achieves a root mean square error (RMSE) of 4.74 dB and a mean absolute percentage error (MAPE) of 8.7% on independent test data. Furthermore, KADL exhibits superior robustness, with a standard deviation of RMSE as low as 0.18 dB across multiple trials. All these results validate the superior accuracy, robustness, and generalization ability of KADL for large-scale backscatter prediction. Full article

Human behavior is shaped by a pervasive motive to align with others, manifesting across a wide range of tendencies—from motor synchrony and emotional contagion to convergence in beliefs and choices. Existing accounts explain how alignment arises through predictive coding and observation–execution mechanisms, but they do not address how it is regulated in a manner that considers when alignment is adaptive and with whom it should occur. We propose a goal-directed model of social alignment that integrates computational and neural levels of analysis, to enhance our understanding of alignment as a context-sensitive decision process rather than a reflexive social tendency. Computationally, alignment is formalized as a prediction-error minimization process over the gap between self and other, augmented by a meta-learning layer in which the learning rate is adaptively tuned according to the inferred value of aligning versus maintaining independence. Assessments of the traits and mental states of self and other serve as key inputs to this regulatory function. Neurally, higher-order representations of these inputs are carried by the mentalizing network (dmPFC, TPJ), which exerts top-down control through the executive control network (dlPFC, rIFG) to enhance or inhibit alignment tendencies generated by observation–execution (mirror) circuitry. By reframing alignment as a form of social decision-making under uncertainty, the model specifies both the computations and neural circuits that integrate contextual cues to arbitrate when and with whom to align. It yields testable predictions across developmental, comparative, cognitive, and neurophysiological domains, and provides a unified framework for understanding the adaptive functions of social alignment, such as strategic social learning, as well as its maladaptive outcomes, including groupthink and false information cascades. Full article

MicroRNAs are closely associated with various physiological and pathological processes, making their in situ fluorescence imaging crucial for functional studies and disease diagnosis. Current methods for the in situ fluorescence imaging of microRNA predominantly rely on linear signal amplification, resulting in relatively weak imaging signals. This study introduces a Y-shaped cascade assembly (YCA) method for high-brightness microRNA imaging in cells. Triggered by target microRNA, catalytic hairpin assembly forms double-stranded DNA (H). Through annealing and hybridization, a Y-shaped structure (P) is created. These components assemble into DNA nanofluorescent particles with multiple FAM fluorophores, significantly amplifying fluorescence signals. Optimization experiments revealed that a 1:1 ratio of P to H and an assembly time of 60 min yielded the best results. Under these optimal conditions, the resulting fluorescent nanoparticles exhibited diameters of 664.133 nm, as observed by DLS. In Huh7 liver cancer cells, YCA generated DNA nanoparticles with a fluorescence intensity increase of 117.77%, triggered by target microRNA-21, producing high-intensity fluorescence images and enabling qualitative detection of microRNA-21. The YCA in situ imaging method offers excellent imaging quality and high efficiency, providing a robust and reliable analytical tool for the diagnosis and monitoring of microRNA-related diseases. Full article

COVID-19 was one of the most serious global public health emergencies in recent years, and its extremely fast spreading speed had a profound negative impact on society. A comprehensive analysis and prediction of COVID-19 could lay a theoretical foundation for monitoring and early warning systems. Since the outbreak of COVID-19, there has been an influx of research on predictive modelling, with artificial intelligence (AI) techniques, particularly machine learning (ML) methods, becoming the dominant research direction due to their superior capability in processing multidimensional datasets and capturing complex nonlinear transmission patterns. We systematically reviewed COVID-19 ML prediction models developed under the background of the epidemic using the PRISMA method. We used the selected keywords to screen the relevant literature of COVID-19 prediction using ML technology from 2020 to 2023 in the Web of Science, Springer and Elsevier databases. Based on predetermined inclusion and exclusion criteria, 136 eligible studies were ultimately selected from 5731 preliminarily screened publications, and the datasets, data preprocessing, ML models, and evaluation metrics used in these studies were assessed. By establishing a multi-level classification framework that included traditional statistical models (such as ARIMA), ML models (such as SVM), deep learning (DL) models (such as CNN, LSTM), ensemble learning methods (such as AdaBoost), and hybrid models (such as the fusion architecture of intelligent optimization algorithms and neural networks), it revealed that the hybrid modelling strategy effectively improved the prediction accuracy of the model through feature combination optimization and model cascade integration. In addition, we compared the performance of ML models with other models in the COVID-19 prediction task. The results showed that the propagation of COVID-19 is affected by multiple factors, including meteorological and socio-economic conditions. Compared to traditional methods, ML methods demonstrated significant advantages in COVID-19 prediction, especially hybrid modelling strategies, which showed great potential in optimizing accuracy. However, these techniques face challenges and limitations despite their strong performance. By reviewing existing research on COVID-19 prediction, this study provided systematic theoretical support for AI applications in infectious disease prediction and promoted technological innovation in public health. Full article

The increasing density and complexity of electromagnetic signals have brought new challenges to multi-component radar signal recognition. To address the problem of low recognition accuracy under low signal-to-noise ratios (SNR) in adapting the common recognition framework of combining time–frequency transformations (TFTs) with convolutional neural networks (CNNs), this paper proposes a new dual-component radar signal recognition framework (TFGM-RMNet) that combines a deep time–frequency generation module with a Transformer-based residual network. First, the received noisy signal is preprocessed. Then, the deep time–frequency generation module is used to learn the complete basis function to obtain various TF features of the time signal, and the corresponding time–frequency representation (TFR) is output under the supervision of high-quality images. Next, a ResNet combined with cascaded multi-head attention (MHSA) is applied to extract local and global features from the TFR. Finally, modulation format prediction is achieved through multi-label classification. The proposed framework does not require explicit TFT during testing, and the TFT process is built into TFGM to replace the traditional TFT. The classification results and ideal TFR are obtained during testing, realizing an end-to-end deep learning (DL) framework. The simulation results show that, when SNR > −8 dB, this method can achieve an average recognition accuracy close to 100%. It achieves 97% accuracy even at an SNR of −10 dB. At the same time, under low SNR, the recognition performance is better than the existing algorithms including DCNN-RAMIML, DCNN-MLL, and DCNN-MIML. Full article

Permeable dams are an important means for river management and ecology protection. Reasonable dam spacing will help regulate sediment transport and reduce sediment load in lakes. Flume experiments were carried out to investigate the effects of hydrological sediment conditions and dam spacing on sediment retention performance and permeability of the cascade permeable dams. The experimental results show that the permeability coefficient of the 1# dam decreased by about 30–40% with a large rate during the initial experiment stage. The decrease amplitude in the permeability coefficient and rising rate of the water level in front of the 1# dam for a large dam spacing (D/L) are positively correlated with the flow rate. At D/L = 5, the water level difference of 1# dam at the end of the experiment was significantly higher than that of other spacing. The sediment mass retained by 1# dam accounts for about 41–65% of the total sediment mass retained, which is about twice that of 2# dam, and plays a major role in cascade permeable dams. A mathematical model for predicting the spatial-temporal sediment concentration inside 1# dam is proposed based on the seepage theory of porous media. The research results are of great guiding significance for the design of the dam parameters. Full article

In recent years, synthetic Computed Tomography (CT) images generated from Magnetic Resonance (MR) or Cone Beam Computed Tomography (CBCT) acquisitions have been shown to be comparable to real CT images in terms of dose computation for radiotherapy simulation. However, until now, there has been no independent strategy to assess the quality of each synthetic image in the absence of ground truth. In this work, we propose a Deep Learning (DL)-based framework to predict the accuracy of synthetic CT in terms of Mean Absolute Error (MAE) without the need for a ground truth (GT). The proposed algorithm generates a volumetric map as an output, informing clinicians of the predicted MAE slice-by-slice. A cascading multi-model architecture was used to deal with the complexity of the MAE prediction task. The workflow was trained and tested on two cohorts of head and neck cancer patients with different imaging modalities: 27 MR scans and 33 CBCT. The algorithm evaluation revealed an accurate HU prediction (a median absolute prediction deviation equal to 4 HU for CBCT-based synthetic CTs and 6 HU for MR-based synthetic CTs), with discrepancies that do not affect the clinical decisions made on the basis of the proposed estimation. The workflow exhibited no systematic error in MAE prediction. This work represents a proof of concept about the feasibility of synthetic CT evaluation in daily clinical practice, and it paves the way for future patient-specific quality assessment strategies. Full article

HLA-B27 and HLA-B57 are associated with autoimmunity and long-term viral control and protection against HIV and HCV infection; however, their role in cancer immunity remains unknown. HLA class I molecules interact with innate checkpoint receptors of the LILRA, LILRB and KIR families present in diverse sets of immune cells. Here, we demonstrate that an open format (peptide free conformation) and expression- and stability-optimized HLA-B57-B2m-IgG4_Fc fusion protein (IOS-1002) binds to human leukocyte immunoglobulin-like receptor B1 and B2 (LILRB1 and LILRB2) and to killer immunoglobulin-like receptor 3DL1 (KIR3DL1). In addition, we show that the IgG4 Fc backbone is required for engagement to Fcγ receptors and potent activation of macrophage phagocytosis. IOS-1002 blocks the immunosuppressive ITIM and SHP1/2 phosphatase signaling cascade, reduces the expression of immunosuppressive M2-like polarization markers of macrophages and differentiation of monocytes to myeloid-derived suppressor cells, enhances tumor cell phagocytosis in vitro and potentiates activation of T and NK cells. Lastly, IOS-1002 demonstrates efficacy in an ex vivo patient-derived tumor sample tumoroid model. IOS-1002 is a first-in-class multi-target and multi-functional human-derived HLA molecule that activates anti-tumor immunity and is currently under clinical evaluation. Full article

Geological mapping involves the identification of elements such as rocks, soils, and surface water, which are fundamental tasks in Geological Environment Remote Sensing (GERS) interpretation. High-precision intelligent interpretation technology can not only reduce labor requirements and significantly improve the efficiency of geological mapping but also assist geological disaster prevention assessment and resource exploration. However, the high interclass similarity, high intraclass variability, gradational boundaries, and complex distributional characteristics of GERS elements coupled with the difficulty of manual labeling and the interference of imaging noise, all limit the accuracy of DL-based methods in wide-area GERS interpretation. We propose a Transformer-based multi-stage and multi-scale fusion network, RSWFormer (Rock–Soil–Water Network with Transformer), for geological mapping of spatially large areas. RSWFormer first uses a Multi-stage Geosemantic Hierarchical Sampling (MGHS) module to extract geological information and high-dimensional features at different scales from local to global, and then uses a Multi-scale Geological Context Enhancement (MGCE) module to fuse geological semantic information at different scales to enhance the understanding of contextual semantics. The cascade of the two modules is designed to enhance the interpretation and performance of GERS elements in geologically complex areas. The high mountainous and hilly areas located in western China were selected as the research area. A multi-source geological remote sensing dataset containing diverse GERS feature categories and complex lithological characteristics, Multi-GL9, is constructed to fill the significant gaps in the datasets required for extensive GERS. Using overall accuracy as the evaluation index, RSWFormer achieves 92.15% and 80.23% on the Gaofen-2 and Landsat-8 datasets, respectively, surpassing existing methods. Experiments show that RSWFormer has excellent performance and wide applicability in geological mapping tasks. Full article

This paper proposes a deep learning (DL)-assisted high-pass-filter (HPF)-based fixed-threshold decision (FTD) for free-space optical (FSO) communication. HPF is applied to reduce the scintillation effect by filtering out the low-frequency components of the received signal. However, the performance is limited owing to the signal distortion from HPF and remnant scintillation effect due to insufficient filtering. Therefore, the DL model is adopted to improve the performance of HPF-based scintillation effect compensation. The multilayer perceptron (MLP) model is used to adaptively select the peak frequency component of the received signal as the optimized cutoff frequency of HPF. Furthermore, recurrent neural network (RNN) and long short-term memory (LSTM) models are cascaded after HPF to compensate for the remnant scintillation effect and recover the signal distortion without the optimization of HPF cutoff frequency. The simulation was conducted under different turbulence channels and data rates. Simulation results showed that MLP-assisted adaptive optimized cutoff frequency and cascaded LSTM and HPF methods were close to the adaptive-threshold decision with precise channel state information under various turbulence channel degrees. Full article

Cancer immunotherapy is a rapidly developing field of medicine that aims to use the host’s immune mechanisms to inhibit and eliminate cancer cells. Antibodies targeting CTLA-4, PD-1, and its ligand PD-L1 are used in various cancer therapies. However, the most thoroughly researched pathway targeting PD-1/PD-L1 has many limitations, and multiple malignancies resist its effects. Human endogenous retrovirus-H Long repeat-associating 2 (HHLA2, known as B7H5/B7H7/B7y) is the youngest known molecule from the B7 family. HHLA2/TMIGD2/KIRD3DL3 is one of the critical pathways in modulating the immune response. Recent studies have demonstrated that HHLA2 has a double effect in modulating the immune system. The connection of HHLA2 with TMIGD2 induces T cell growth and cytokine production via an AKT-dependent signaling cascade. On the other hand, the binding of HHLA2 and KIR3DL3 leads to the inhibition of T cells and mediates tumor resistance against NK cells. This review aimed to summarize novel information about HHLA2, focusing on immunological mechanisms and clinical features of the HHLA2/KIR3DL3/TMIGD2 pathway in the context of potential strategies for malignancy treatment. Full article

Glutamate is a major excitatory neurotransmitter that mediates neuronal damage in acute and chronic brain disorders. The effect and mechanism of phillygenin, a natural compound with neuroprotective potential, on glutamate release in isolated nerve terminals (synaptosomes) prepared from the rat cerebral cortex were examined. In this study, 4-aminopyridine (4-AP), a potassium channel blocker, was utilized to induce the release of glutamate, which was subsequently quantified via a fluorometric assay. Our findings revealed that phillygenin reduced 4-AP-induced glutamate release, and this inhibitory effect was reversed by removing extracellular Ca²⁺ or inhibiting vesicular transport with bafilomycin A1. However, exposure to the glutamate transporter inhibitor dl-threo-beta-benzyl-oxyaspartate (dl-TOBA) did not influence the inhibitory effect. Moreover, phillygenin did not change the synaptosomal membrane potential but lowered the 4-AP-triggered increase in intrasynaptosomal Ca²⁺ concentration ([Ca²⁺]_i). Antagonizing Ca_v2.2 (N-type) calcium channels blocked the inhibition of glutamate release by phillygenin, whereas pretreatment with the mitochondrial Na⁺/Ca²⁺ exchanger inhibitor, CGP37157 or the ryanodine receptor inhibitor, dantrolene, both of which block intracellular Ca²⁺ release, had no effect. The effect of phillygenin on glutamate release triggered by 4-AP was completely abolished when MAPK/ERK inhibitors were applied. Furthermore, phillygenin attenuated the phosphorylation of ERK1/2 and its major presynaptic target, synapsin I, a protein associated with synaptic vesicles. These data collectively suggest that phillygenin mediates the inhibition of evoked glutamate release from synaptosomes primarily by reducing the influx of Ca²⁺ through Ca_v2.2 calcium channels, thereby subsequently suppressing the MAPK/ERK/synapsin I signaling cascade. Full article

In this study, we explore the integration of cascading and ensemble techniques in Deep Learning (DL) to improve prediction accuracy on diabetes data. The primary approach involves creating multiple Neural Networks (NNs), each predicting the outcome independently, and then feeding these initial predictions into another set of NN. Our exploration starts from an initial preliminary study and extends to various ensemble techniques including bagging, stacking, and finally cascading. The cascading ensemble involves training a second layer of models on the predictions of the first. This cascading structure, combined with ensemble voting for the final prediction, aims to exploit the strengths of multiple models while mitigating their individual weaknesses. Our results demonstrate significant improvement in prediction accuracy, providing a compelling case for the potential utility of these techniques in healthcare applications, specifically for prediction of diabetes where we achieve compelling model accuracy of 91.5% on the test set on a particular challenging dataset, where we compare thoroughly against many other methodologies. Full article