Association between frailty and acute kidney injury after cardiac surgery: unraveling the moderation effect of body fat through an international, retrospective, multi-cohort study.

BACKGROUND: Acute kidney injury (AKI) is a common and serious complication after cardiac surgery that significantly affects patient outcomes. Given the limited treatment options available, identifying modifiable risk factors is critical. Frailty and obesity, two heterogeneous physiological states, have significant implications for identifying and preventing AKI. Our study investigated the interplay among frailty, body composition, and AKI risk after cardiac surgery to inform patient management strategies. MATERIAL AND METHODS: This retrospective cohort study included three international cohorts. Primary analysis was conducted in adult patients who underwent cardiac surgery between 2014 and 2019 at Wuhan XX Hospital, China. We tested the generalizability of our findings with data from two independent international cohorts, the Medical Information Mart for Intensive Care IV (MIMIC-IV) and the eICU Collaborative Research Database. Frailty was assessed using a clinical lab-based frailty index (FI-LAB), while total body fat percentage (BF%) was calculated based on a formula accounting for BMI, sex, and age. Logistic regression models were used to analyze the associations between frailty, body fat, and AKI, adjusting for pertinent covariates. RESULTS: A total of 8785 patients across three international cohorts were included in the study. In the primary analysis of 3,569 patients from Wuhan XX Hospital, moderate and severe frailty were associated with an increased AKI risk after cardiac surgery. Moreover, a nonlinear relationship was observed between body fat percentage and AKI risk. When stratified by the degree of frailty, lower body fat correlated with a decreased incidence of AKI. Extended analyses using the MIMIC-IV and eICU cohorts (n=3,951 and n=1,265, respectively) validated these findings and demonstrated that a lower total BF% was associated with decreased AKI incidence. Moderation analysis revealed that the effect of frailty on AKI risk was moderated by the body fat percentage. Sensitivity analyses demonstrated results consistent with the main analyses. CONCLUSION: Higher degrees of frailty were associated with an elevated risk of AKI following cardiac surgery, and total BF% moderated this relationship. This research underscores the significance of integrating frailty and body fat assessments into routine cardiovascular care to identify high-risk patients for AKI and implement personalized interventions to improve patient outcomes.

Epileptic Seizure Prediction Using Spatiotemporal Feature Fusion on EEG.

Electroencephalography (EEG) plays a crucial role in epilepsy analysis, and epileptic seizure prediction has significant value for clinical treatment of epilepsy. Currently, prediction methods using Convolutional Neural Network (CNN) primarily focus on local features of EEG, making it challenging to simultaneously capture the spatial and temporal features from multi-channel EEGs to identify the preictal state effectively. In order to extract inherent spatial relationships among multi-channel EEGs while obtaining their temporal correlations, this study proposed an end-to-end model for the prediction of epileptic seizures by incorporating Graph Attention Network (GAT) and Temporal Convolutional Network (TCN). Low-pass filtered EEG signals were fed into the GAT module for EEG spatial feature extraction, and followed by TCN to capture temporal features, allowing the end-to-end model to acquire the spatiotemporal correlations of multi-channel EEGs. The system was evaluated on the publicly available CHB-MIT database, yielding segment-based accuracy of 98.71%, specificity of 98.35%, sensitivity of 99.07%, and F1-score of 98.71%, respectively. Event-based sensitivity of 97.03% and False Positive Rate (FPR) of 0.03/h was also achieved. Experimental results demonstrated this system can achieve superior performance for seizure prediction by leveraging the fusion of EEG spatiotemporal features without the need of feature engineering.

Human attention guided explainable artificial intelligence for computer vision models.

Explainable artificial intelligence (XAI) has been increasingly investigated to enhance the transparency of black-box artificial intelligence models, promoting better user understanding and trust. Developing an XAI that is faithful to models and plausible to users is both a necessity and a challenge. This work examines whether embedding human attention knowledge into saliency-based XAI methods for computer vision models could enhance their plausibility and faithfulness. Two novel XAI methods for object detection models, namely FullGrad-CAM and FullGrad-CAM++, were first developed to generate object-specific explanations by extending the current gradient-based XAI methods for image classification models. Using human attention as the objective plausibility measure, these methods achieve higher explanation plausibility. Interestingly, all current XAI methods when applied to object detection models generally produce saliency maps that are less faithful to the model than human attention maps from the same object detection task. Accordingly, human attention-guided XAI (HAG-XAI) was proposed to learn from human attention how to best combine explanatory information from the models to enhance explanation plausibility by using trainable activation functions and smoothing kernels to maximize the similarity between XAI saliency map and human attention map. The proposed XAI methods were evaluated on widely used BDD-100K, MS-COCO, and ImageNet datasets and compared with typical gradient-based and perturbation-based XAI methods. Results suggest that HAG-XAI enhanced explanation plausibility and user trust at the expense of faithfulness for image classification models, and it enhanced plausibility, faithfulness, and user trust simultaneously and outperformed existing state-of-the-art XAI methods for object detection models.

Cosine convolutional neural network and its application for seizure detection.

Traditional convolutional neural networks (CNNs) often suffer from high memory consumption and redundancy in their kernel representations, leading to overfitting problems and limiting their application in real-time, low-power scenarios such as seizure detection systems. In this work, a novel cosine convolutional neural network (CosCNN), which replaces traditional kernels with the robust cosine kernel modulated by only two learnable factors, is presented, and its effectiveness is validated on the tasks of seizure detection. Meanwhile, based on the cosine lookup table and KL-divergence, an effective post-training quantization algorithm is proposed for CosCNN hardware implementation. With quantization, CosCNN can achieve a nearly 75% reduction in the memory cost with almost no accuracy loss. Moreover, we design a configurable cosine convolution accelerator on Field Programmable Gate Array (FPGA) and deploy the quantized CosCNN on Zedboard, proving the proposed seizure detection system can operate in real-time and low-power scenarios. Extensive experiments and comparisons were conducted using two publicly available epileptic EEG databases, the Bonn database and the CHB-MIT database. The results highlight the performance superiority of the CosCNN over traditional CNNs as well as other seizure detection methods.

Nickel selenide/g-C3N4 heterojunction photocatalyst promotes CC coupling for photocatalytic CO2 reduction to ethane.

Photocatalytic CO2 reduction to generate high value-added and renewable chemicals is of great potential in facilitating the realization of closed-loop and carbon-neutral hydrogen economy. Stabilizing and accelerating the formation of COCO* intermediate is crucial to achieve high-selectivity ethane production. Herein, a novel 3D/2D NiSe2/g-C3N4 heterostructure that mesoscale hedgehog nickel selenide (NiSe2) grown on the ultrathin g-C3N4 nanosheets were synthesized via a successively high temperature calcination process and in-situ thermal injection method for the first time. The optimum 2.7 % NiSe2/g-C3N4 heterostructure achieved moderate C2H6 generation rate of 46.1 µmol·g-1·h-1 and selectivity of 97.5 % without any additional photosensitizers and sacrificial agents under light illumination. Based on the results of the theoretical calculations and experiments, the improvement of photocatalytic CO2 to C2H6 production and selectivity should be ascribed to the increased visible light absorption ability, unique 3D/2D heterostructures with promoted adsorption of CO2 molecules on the Ni active sites, the type II heterojunction with improved charge transfer dynamics and lowered interfacial transfer resistance, as well as the formation of COCO* key intermediate. This work provides an inspiration to construct efficient photocatalysts for the direct transformation of CO2 to multicarbon products (C2+).

Remote ischemic preconditioning reduces mitochondrial apoptosis mediated by calpain 1 activation in myocardial ischemia-reperfusion injury through calcium channel subunit Cacna2d3.

Remote Ischemic Preconditioning (RIPC) can reduce myocardial ischemia-reperfusion injury, but its mechanism is not clear. In order to explore the mechanism of RIPC in myocardial protection, we collected myocardial specimens during cardiac surgery in children with tetralogy of Fallot for sequencing. Our study found RIPC reduces the expression of the calcium channel subunit cacna2d3, thereby impacting the function of calcium channels. As a result, calcium overload during ischemia-reperfusion is reduced, and the activation of calpain 1 is inhibited. This ultimately leads to a decrease in calpain 1 cleavage of Bax, consequently inhibiting increased mitochondrial permeability-mediated apoptosis. Notably, in both murine and human models of myocardial ischemia-reperfusion injury, RIPC inhibiting the expression of the calcium channel subunit cacna2d3 and the activation of calpain 1, improving cardiac function and histological outcomes. Overall, our findings put forth a proposed mechanism that elucidates how RIPC reduces myocardial ischemia-reperfusion injury, ultimately providing a solid theoretical foundation for the widespread clinic application of RIPC.

MSGF-GLP: fusion method of visible and hyperspectral data for early detection of discolored standing trees.

Pest and disease damage to forests cannot be underestimated, so it is essential to detect diseased trees in time and take measures to stop their spread. The detection of discoloration standing trees is one of the important means to effectively control the spread of pests and diseases. In the visible wavelength range, early infected trees do not show significant color changes, which poses a challenge for early detection and is only suitable for monitoring middle and late discolor trees. The spectral resolution of hyperspectral restricts the improvement of its spatial resolution, and there are phenomena of different spectral of the same and foreign objects in the same spectrum, which affect the detection results. In this paper, the method of hyperspectral and CCD image fusion is used to achieve high-precision detection of discoloration standing trees. This paper proposes an improved algorithm MSGF-GLP, which uses multi-scale detail boosting and MTF filter to refine high-resolution data. By combining guided filtering with hyperspectral images, the spatial detail difference is enhanced, and the injection gain is interpolated into the difference of each band, so as to obtain high-resolution and high-quality hyperspectral images. This research is based on hyperspectral and CCD data obtained from LiCHy, Chinese Academy of Forestry, Maoershan Experimental Forest Farm, Shangzhi City, Heilongjiang Province. The evaluation framework is used to compare with the other five fusion algorithms to verify the good effect of the proposed method, which can effectively preserve the canopy spectrum and improve the spatial details. The fusion results of forestry remote sensing data were analyzed using the vegetation Normalized Difference Water Index and Plant Senescence Reflectance Index. The fused results can be used to distinguish the difference between discoloration trees and healthy trees by the multispectral vegetation index. The research results can provide good technical support for the practical application of forest remote sensing data fusion, and lay the foundation for promoting the scientific, automatic and intelligent forestry control.

Self-attention-based convolutional neural network and time-frequency common spatial pattern for enhanced motor imagery classification.

BACKGROUND: Motor imagery (MI) based brain-computer interfaces (BCIs) have promising potentials in the field of neuro-rehabilitation. However, due to individual variations in active brain regions during MI tasks, the challenge of decoding MI EEG signals necessitates improved classification performance for practical application. NEW METHOD: This study proposes a self-attention-based Convolutional Neural Network (CNN) in conjunction with a time-frequency common spatial pattern (TFCSP) for enhanced MI classification. Due to the limited availability of training data, a data augmentation strategy is employed to expand the scale of MI EEG datasets. The self-attention-based CNN is trained to automatically extract the temporal and spatial information from EEG signals, allowing the self-attention module to select active channels by calculating EEG channel weights. TFCSP is further implemented to extract multiscale time-frequency-space features from EEG data. Finally, the EEG features derived from TFCSP are concatenated with those from the self-attention-based CNN for MI classification. RESULTS: The proposed method is evaluated on two publicly accessible datasets, BCI Competition IV IIa and BCI Competition III IIIa, yielding mean accuracies of 79.28 % and 86.39 %, respectively. CONCLUSIONS: Compared with state-of-the-art methods, our approach achieves superior classification results in accuracy. Self-attention-based CNN combining with TFCSP can make full use of the time-frequency-space information of EEG, and enhance the classification performance.

Calpain: the regulatory point of myocardial ischemia-reperfusion injury.

Calpain is a conserved cysteine protease readily expressed in several mammalian tissues, which is usually activated by Ca2+ and with maximum activity at neutral pH. The activity of calpain is tightly regulated because its aberrant activation will nonspecifically cleave various proteins in cells. Abnormally elevation of Ca2+ promotes the abnormal activation of calpain during myocardial ischemia-reperfusion, resulting in myocardial injury and cardiac dysfunction. In this paper, we mainly reviewed the effects of calpain in various programmed cell death (such as apoptosis, mitochondrial-mediated necrosis, autophagy-dependent cell death, and parthanatos) in myocardial ischemia-reperfusion. In addition, we also discussed the abnormal activation of calpain during myocardial ischemia-reperfusion, the effect of calpain on myocardial repair, and the possible future research directions of calpain.

Automatic Seizure Detection Based on Stockwell Transform and Transformer.

Epilepsy is a chronic neurological disease associated with abnormal neuronal activity in the brain. Seizure detection algorithms are essential in reducing the workload of medical staff reviewing electroencephalogram (EEG) records. In this work, we propose a novel automatic epileptic EEG detection method based on Stockwell transform and Transformer. First, the S-transform is applied to the original EEG segments, acquiring accurate time-frequency representations. Subsequently, the obtained time-frequency matrices are grouped into different EEG rhythm blocks and compressed as vectors in these EEG sub-bands. After that, these feature vectors are fed into the Transformer network for feature selection and classification. Moreover, a series of post-processing methods were introduced to enhance the efficiency of the system. When evaluating the public CHB-MIT database, the proposed algorithm achieved an accuracy of 96.15%, a sensitivity of 96.11%, a specificity of 96.38%, a precision of 96.33%, and an area under the curve (AUC) of 0.98 in segment-based experiments, along with a sensitivity of 96.57%, a false detection rate of 0.38/h, and a delay of 20.62 s in event-based experiments. These outstanding results demonstrate the feasibility of implementing this seizure detection method in future clinical applications.

The Fra-1: Novel role in regulating extensive immune cell states and affecting inflammatory diseases.

Fra-1(Fos-related antigen1), a member of transcription factor activator protein (AP-1), plays an important role in cell proliferation, apoptosis, differentiation, inflammation, oncogenesis and tumor metastasis. Accumulating evidence suggest that the malignancy and invasive ability of tumors can be significantly changed by directly targeting Fra-1. Besides, the effects of Fra-1 are gradually revealed in immune and inflammatory settings, such as arthritis, pneumonia, psoriasis and cardiovascular disease. These regulatory mechanisms that orchestrate immune and non-immune cells underlie Fra-1 as a potential therapeutic target for a variety of human diseases. In this review, we focus on the current knowledge of Fra-1 in immune system, highlighting its unique importance in regulating tissue homeostasis. In addition, we also discuss the possible critical intervention strategy in diseases, which also outline future research and development avenues.

Positive interaction between GPER and ß-alanine in the dorsal root ganglion uncovers potential mechanisms: mediating continuous neuronal sensitization and neuroinflammation responses in neuropathic pain.

BACKGROUND: The pathogenesis of neuropathic pain and the reasons for the prolonged unhealing remain unknown. Increasing evidence suggests that sex oestrogen differences play a role in pain sensitivity, but few studies have focused on the oestrogen receptor which may be an important molecular component contributing to peripheral pain transduction. We aimed to investigate the impact of oestrogen receptors on the nociceptive neuronal response in the dorsal root ganglion (DRG) and spinal dorsal horn using a spared nerve injury (SNI) rat model of chronic pain. METHODS: We intrathecally (i.t.) administered a class of oestrogen receptor antagonists and agonists intrathecal (i.t.) administrated to male rats with SNI or normal rats to identify the main receptor. Moreover, we assessed genes identified through genomic metabolic analysis to determine the key metabolism point and elucidate potential mechanisms mediating continuous neuronal sensitization and neuroinflammatory responses in neuropathic pain. The excitability of DRG neurons was detected using the patch-clamp technique. Primary culture was used to extract microglia and DRG neurons, and siRNA transfection was used to silence receptor protein expression. Immunofluorescence, Western blotting, RT-PCR and behavioural testing were used to assess the expression, cellular distribution, and actions of the main receptor and its related signalling molecules. RESULTS: Increasing the expression and function of G protein-coupled oestrogen receptor (GPER), but not oestrogen receptor-α (ERα) and oestrogen receptor-ß (ERß), in the DRG neuron and microglia, but not the dorsal spinal cord, contributed to SNI-induced neuronal sensitization. Inhibiting GPER expression in the DRG alleviated SNI-induced pain behaviours and neuroinflammation by simultaneously downregulating iNOS, IL-1ß and IL-6 expression and restoring GABAα2 expression. Additionally, the positive interaction between GPER and ß-alanine and subsequent ß-alanine accumulation enhances pain sensation and promotes chronic pain development. CONCLUSION: GPER activation in the DRG induces a positive association between ß-alanine with iNOS, IL-1ß and IL-6 expression and represses GABAα2 involved in post-SNI neuropathic pain development. Blocking GPER and eliminating ß-alanine in the DRG neurons and microglia may prevent neuropathic pain development.

Epithelial­derived exosomes promote M2 macrophage polarization via Notch2/SOCS1 during mechanical ventilation.

Alveolar macrophages (AMs) play an essential role in ventilator­induced lung injury (VILI). Exosomes and their cargo, including microRNAs (miRNAs/miRs) serve as regulators of the intercellular communications between macrophages and epithelial cells (ECs), and are involved in maintaining homeostasis in lung tissue. The present study found that exosomes released by ECs subjected to cyclic stretching promoted M2 macrophage polarization. It was demonstrated that miR­21a­5p, upregulated in epithelial­derived exosomes, increased the percentage of M2 macrophages by suppressing the expression of Notch2 and the suppressor of cytokine signaling 1 (SOCS1). The overexpression of Notch2 decreased the percentage of M2 macrophages. However, these effects were reversed by the downregulation of SOCS1. The percentage of M2 macrophages was increased in both short­term high­ and low­tidal­volume mechanical ventilation, and the administration of exosomes­derived from cyclically stretched ECs had the same function. However, the administration of miR­21a­5p antagomir decreased M2 macrophage activation induced by cyclically stretched ECs or ventilation. Thus, the present study demonstrates that the intercellular transferring of exosomes from ECs to AMs promotes M2 macrophage polarization. Exosomes may prove to be a novel treatment for VILI.

Targeting C-C Chemokine Receptor 5: Key to Opening the Neurorehabilitation Window After Ischemic Stroke.

Stroke is the world's second major cause of adult death and disability, resulting in the destruction of brain tissue and long-term neurological impairment; induction of neuronal plasticity can promote recovery after stroke. C-C chemokine receptor 5 (CCR5) can direct leukocyte migration and localization and is a co-receptor that can mediate human immunodeficiency virus (HIV) entry into cells. Its role in HIV infection and immune response has been extensively studied. Furthermore, CCR5 is widely expressed in the central nervous system (CNS), is engaged in various physiological activities such as brain development, neuronal differentiation, communication, survival, and learning and memory capabilities, and is also involved in the development of numerous neurological diseases. CCR5 is differentially upregulated in neurons after stroke, and the inhibition of CCR5 in specific regions of the brain promotes motor and cognitive recovery. The mechanism by which CCR5 acts as a therapeutic target to promote neurorehabilitation after stroke has rarely been systematically reported yet. Thus, this review aims to discuss the function of CCR5 in the CNS and the mechanism of its effect on post-stroke recovery by regulating neuroplasticity and the inflammatory response to provide an effective basis for clinical rehabilitation after stroke.

Multiscale time-frequency method for multiclass Motor Imagery Brain Computer Interface.

Motor Imagery Brain Computer Interface (MI-BCI) has become a promising technology in the field of neurorehabilitation. However, the performance and computational complexity of the current multiclass MI-BCI have not been fully optimized, and the intuitive interpretation of individual differences on motor imagery tasks is seldom investigated. In this paper, a well-designed multiscale time-frequency segmentation scheme is first applied to multichannel EEG recordings to obtain Time-Frequency Segments (TFSs). Then, the TFS selection based on a specific wrapper feature selection rule is utilized to determine optimum TFSs. Next, One-Versus-One (OvO)-divCSP implemented in divergence framework is used to extract discriminative features. Finally, One-Versus-Rest (OvR)-SVM is utilized to predict the class label based on selected multiclass MI features. Experimental results indicate our method yields a superior performance on two publicly available multiclass MI datasets with a mean accuracy of 80.00% and a mean kappa of 0.73. Meanwhile, the proposed TFS selection method can significantly alleviate the computational burden with little accuracy reduction, demonstrating the feasibility of real-time multiclass MI-BCI. Furthermore, the Motor Imagery Time-Frequency Reaction Map (MI-TFRM) is visualized, contributing to analyzing and interpreting the performance differences between different subjects.

Mechanochemical coordination self-assembly for Cobalt-based metal-organic framework-derived bifunctional oxygen electrocatalysts.

The exploration and preparation of inexpensive, high-performance and stable catalysts for oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) is of significant imperativeness, yet is still on the way. In this study, a facile operation protocol is presented for constructing an exquisite three-dimensional coral reef-like carbon nanotube assembly bridged with N-doped graphene (assigned as 3D CNTAs-NG, which represented carbonization products at 900℃) as highly efficient and durable ORR/OER electrocatalysts. It does not require the introduction of reductive atmosphere. In this tactic, the dicyanamide ligand on the Co-MOF not only was instrumental in the introduction of nitrogen but also acted as the inducer of carbon nanotubes (CNTs) to lock the metallic Co in graphitic carbon layers. Graphene oxide (GO) is chosen as a matrix to pin the CNTs and ensure the uniform distribution of CNTs. The obtained CNTAs-NG structure possesses 3D open porous texture, abundant defects, desired nitrogen bonding type and high specific surface area, providing them with excellent ORR and OER properties. As such, the optimized 3D CNTAs-NG sample shows high onset potential (Eonset = 0.97 V vs. RHE) and half-wave potential (E1/2 = 0.85 V vs. RHE) for ORR and overpotential of 340 mV at 10 mA∙cm-2 for OER. Meanwhile, the prepared optimum catalyst exhibited outstanding durability for ORR and OER in alkaline solutions. This work may pave significant concepts for the synthesis of highly active bifunctional electrocatalysts with intriguing architectures and compositions.

Patient-Independent Seizure Detection Based on Channel-Perturbation Convolutional Neural Network and Bidirectional Long Short-Term Memory.

Automatic seizure detection is of great significance for epilepsy diagnosis and alleviating the massive burden caused by manual inspection of long-term EEG. At present, most seizure detection methods are highly patient-dependent and have poor generalization performance. In this study, a novel patient-independent approach is proposed to effectively detect seizure onsets. First, the multi-channel EEG recordings are preprocessed by wavelet decomposition. Then, the Convolutional Neural Network (CNN) with proper depth works as an EEG feature extractor. Next, the obtained features are fed into a Bidirectional Long Short-Term Memory (BiLSTM) network to further capture the temporal variation characteristics. Finally, aiming to reduce the false detection rate (FDR) and improve the sensitivity, the postprocessing, including smoothing and collar, is performed on the outputs of the model. During the training stage, a novel channel perturbation technique is introduced to enhance the model generalization ability. The proposed approach is comprehensively evaluated on the CHB-MIT public scalp EEG database as well as a more challenging SH-SDU scalp EEG database we collected. Segment-based average accuracies of 97.51% and 93.70%, event-based average sensitivities of 86.51% and 89.89%, and average AUC-ROC of 90.82% and 90.75% are yielded on the CHB-MIT database and SH-SDU database, respectively.

ECG quality assessment based on hand-crafted statistics and deep-learned S-transform spectrogram features.

Background and Objective Electrocardiogram (ECG) quality assessment is significant for automatic diagnosis of cardiovascular disease and reducing the massive workload of reviewing continuous ECGs. Hence, how to design an appropriate algorithm for objectively evaluating the multi-lead ECG recordings is particularly important. Despite the deep learning methods performing well in many fields, as a data-driven method, it may not be entirely suitable for ECG analysis due to the difficulty in obtaining sufficient data and the low signal-to-noise ratio of ECG recordings. In this study, with the aim of providing an accurate and automatic ECG quality assessment scheme, we propose an innovative ECG quality assessment algorithm based on hand-crafted statistical features and deep-learned spectral features. Methods In this paper, a novel approach, combining the deep-learned Stockwell transform (S-Transform) spectrogram features and hand-crafted statistical features, is proposed for ECG quality assessment. Firstly, a double-input convolutional neural network (CNN) is established. Then, the S-Transform with a novel online augmentation scheme is performed on the multi-lead raw ECG signal received from one input layer to obtain proper time-frequency representation. After that, the CNN with three convolutional layers is employed to extract robust deep-learned features automatically. Simultaneously, the hand-crafted statistical features, including lead-fall, baseline drift, and R peak features, are calculated and fed into another input layer for feature fusion training. Finally, the deep-learned and hand-crafted features are concatenated and further fused by a fully connected layer for quality classification. Furthermore, a log-odds analysis scheme combining with a gradient-based method can localize the abnormal zone in time, frequency, and spatial domains. Results and Conclusion Our proposed method is evaluated on a publicly available database with 10-fold cross-validation. The experimental results demonstrate that the proposed assessment algorithm reached a mean accuracy of 93.09%, a mean F1-score of 0.8472, and a sensitivity of 0.9767. Moreover, comprehensive experiments indicate that the fusion of CNN features and statistical features has complementary advantages and ideal interpretability, achieving end-to-end multi-lead ECG assessment with satisfying performance.

An Efficient and Accurate Iris Recognition Algorithm Based on a Novel Condensed 2-ch Deep Convolutional Neural Network.

Recently, deep learning approaches, especially convolutional neural networks (CNNs), have attracted extensive attention in iris recognition. Though CNN-based approaches realize automatic feature extraction and achieve outstanding performance, they usually require more training samples and higher computational complexity than the classic methods. This work focuses on training a novel condensed 2-channel (2-ch) CNN with few training samples for efficient and accurate iris identification and verification. A multi-branch CNN with three well-designed online augmentation schemes and radial attention layers is first proposed as a high-performance basic iris classifier. Then, both branch pruning and channel pruning are achieved by analyzing the weight distribution of the model. Finally, fast finetuning is optionally applied, which can significantly improve the performance of the pruned CNN while alleviating the computational burden. In addition, we further investigate the encoding ability of 2-ch CNN and propose an efficient iris recognition scheme suitable for large database application scenarios. Moreover, the gradient-based analysis results indicate that the proposed algorithm is robust to various image contaminations. We comprehensively evaluated our algorithm on three publicly available iris databases for which the results proved satisfactory for real-time iris recognition.

Effects of moisture content and tillage methods on creep properties of paddy soil.

The rheological properties parameters of paddy soil affect the interaction between the tillage tools and soil, thus influencing the operation quality and power consumption. In order to study the effects of tillage methods and moisture content on the rheological properties parameters of paddy soil in the middle and lower reaches of the Yangtze River, uniaxial compression creep tests of paddy soils with four moisture contents under no tillage (moisture contents: 26.71%, 24.52%, 23.26%, 21.28%) and plough tillage (moisture contents: 26.77%, 25.55%, 23.40%, 20.56%) were carried out using a TMS-PRO texture analyzer. The creep properties curves obtained from the tests, and the rheological constitutive equation of paddy soil under compression was established by Burgers viscoelastic model. Respectively, the quantitative change rules of creep properties of paddy soil with different moisture contents under different tillage methods and the correlation between these parameters were explored. The results showed that the moisture content under the three-year plough tillage and no tillage methods had significant influence on the rheological properties parameters of paddy soil (P < 0.05). The instantaneous elastic modulus, delay elastic modulus, and viscosity coefficient of the two paddy soils (no tillage and plough tillage soils) decreased with the increase of moisture content. However, the variation rules of relaxation time and delay viscosity coefficient with moisture content differed between these two paddy soils. Specifically, the strain rate of the two paddy soils decreased as moisture content decreased, where the total strain combines elastic strain, viscous strain, and viscoelastic strain. The initial strain rate and steady strain rate of the plough tillage paddy soils were lower than that of the no tillage paddy soils. The established creep model equation could be used to obtain viscoelastic rheological parameters of paddy soil in a wide range. The fitting equations between rheological parameters and moisture content were introduced into Burgers model, and the coupling equations between creep deformation and moisture content and time were derived, which could be used to predict the creep properties and deformation behavior of paddy soil in a certain range of no tillage and ploughed field. Overall, this study has a certain theoretical significance for the development and improvement of paddy soil rheology theory, and can also provide theoretical basis and technical support for the research of agricultural machinery design optimization, field water, soil conservation, soil tillage and compaction related simulation analysis in the middle and lower reaches of the Yangtze River.