Efficacy of auxetic lattice structured shoe sole in advancing footwear comfort-From the perspective of plantar pressure and contact area.

Introduction: Designing footwear for comfort is vital for preventing foot injuries and promoting foot health. This study explores the impact of auxetic structured shoe soles on plantar biomechanics and comfort, motivated by the integration of 3D printing in footwear production and the superior mechanical properties of auxetic designs. The shoe sole designs proposed in this study are based on a three-dimensional re-entrant auxetic lattice structure, orthogonally composed of re-entrant hexagonal honeycombs with internal angles less than 90 degrees. Materials fabricated using this lattice structure exhibit the characteristic of a negative Poisson's ratio, displaying lateral expansion under tension and densification under compression. Methods: The study conducted a comparative experiment among three different lattice structured (auxetic 60°, auxetic 75° and non-auxetic 90°) thermoplastic polyurethane (TPU) shoe soles and conventional polyurethane (PU) shoe sole through pedobarographic measurements and comfort rating under walking and running conditions. The study obtained peak plantar pressures (PPPs) and contact area across seven plantar regions of each shoe sole and analyzed the correlation between these biomechanical parameters and subjective comfort. Results: Compared to non-auxetic shoe soles, auxetic structured shoe soles reduced PPPs across various foot regions and increased contact area. The Auxetic 60°, which had the highest comfort ratings, significantly lowered peak pressures and increased contact area compared to PU shoe sole. Correlation analysis showed that peak pressures in specific foot regions (hallux, second metatarsal head, and hindfoot when walking; second metatarsal head, third to fifth metatarsal head, midfoot, and hindfoot when running) were related to comfort. Furthermore, the contact area in all foot regions was significantly associated with comfort, regardless of the motion states. Conclusion: The pressure-relief performance and conformability of the auxetic lattice structure in the shoe sole contribute to enhancing footwear comfort. The insights provided guide designers in developing footwear focused on foot health and comfort using auxetic structures.

Natural polysaccharides as promising reno-protective agents for the treatment of various kidney injury.

Renal injury, a prevalent clinical outcome with multifactorial etiology, imposes a substantial burden on society. Currently, there remains a lack of effective management and treatments. Extensive research has emphasized the diverse biological effects of natural polysaccharides, which exhibit promising potential for mitigating renal damage. This review commences with the pathogenesis of four common renal diseases and the shared mechanisms underlying renal injury. The renoprotective roles of polysaccharides in vivo and in vitro are summarized in the following five aspects: anti-oxidative stress effects, anti-apoptotic effects, anti-inflammatory effects, anti-fibrotic effects, and gut modulatory effects. Furthermore, we explore the structure-activity relationship and bioavailability of polysaccharides in relation to renal injury, as well as investigate their utility as biomaterials for alleviating renal injury. The clinical experiments of polysaccharides applied to patients with chronic kidney disease are also reviewed. Broadly, this review provides a comprehensive perspective on the research direction of natural polysaccharides in the context of renal injury, with the primary aim to serve as a reference for the clinical development of polysaccharides as pharmaceuticals and prebiotics for the treatment of kidney diseases.

STCS-Net: a medical image segmentation network that fully utilizes multi-scale information.

In recent years, significant progress has been made in the field of medical image segmentation through the application of deep learning and neural networks. Numerous studies have focused on optimizing encoders to extract more comprehensive key information. However, the importance of decoders in directly influencing the final output of images cannot be overstated. The ability of decoders to effectively leverage diverse information and further refine crucial details is of paramount importance. This paper proposes a medical image segmentation architecture named STCS-Net. The designed decoder in STCS-Net facilitates multi-scale filtering and correction of information from the encoder, thereby enhancing the accuracy of extracting vital features. Additionally, an information enhancement module is introduced in skip connections to highlight essential features and improve the inter-layer information interaction capabilities. Comprehensive evaluations on the ISIC2016, ISIC2018, and Lung datasets validate the superiority of STCS-Net across different scenarios. Experimental results demonstrate the outstanding performance of STCS-Net on all three datasets. Comparative experiments highlight the advantages of our proposed network in terms of accuracy and parameter efficiency. Ablation studies confirm the effectiveness of the introduced decoder and skip connection module. This research introduces a novel approach to the field of medical image segmentation, providing new perspectives and solutions for future developments in medical image processing and analysis.

In Situ Growth of Amorphous MnO2 on Graphite Felt via Mild Etching Engineering as a Powerful Catalyst for Advanced Vanadium Redox Flow Batteries.

Owing to the advantages of low cost, high safety, and a desirable cycling lifetime, vanadium redox flow batteries (VRFBs) have attracted great attention in the large-scale energy storage field. However, graphite felts (GFs), widely used as electrode materials, usually possess an inferior catalytic activity for the redox reaction of vanadium ions, largely limiting the energy efficiency and rate performance of VRFBs. Here, an in situ growth of amorphous MnO2 on graphite felt (AMO@GF) was designed for application in VRFBs via mild and rapid etching engineering (5 min). After the etching process, the graphite felt fibers showed a porous and defective surface, contributing to abundant active sites toward the redox reaction. In addition, formed amorphous MnO2 can also serve as a powerful catalyst to facilitate the redox couples of VO2+/VO2+ based on density functional theoretical (DFT) calculations. As a result, the VRFB using AMO@GF displayed an elevated energy efficiency and superior stability after 2400 cycles at 200 mA cm-2, and the maximum current density can reach 300 mA cm-2. Such a high-efficiency and convenient design strategy for the electrode material will drive the further development and industrial application of VRFBs and other flow battery systems.

Impacts of futile reperfusion and reperfusion injury in acute ischemic stroke.

Acute ischemic stroke (AIS) remains to be a challenging cerebrovascular disease. The mainstay of AIS management is endovascular reperfusion therapy, including thrombectomy and thrombolysis. However, ineffective (futile) reperfusion (FR) or reperfusion injury (RI) can be seen in a significant number of patients undergoing reperfusion strategy. In this article, we discuss two clinically relevant concepts known as "time window" and "tissue window" that can impact the clinical outcome of reperfusion therapy. We also explore patient risk factors, leading to FR and RI as well as an emerging concept of "no-reflow phenomenon" seen in ineffective reperfusion. These fundamental concepts provide insight into the clinical management of AIS patients and provide references for future research.

NYUS.2: an automated machine learning prediction model for the large-scale real-time simulation of grapevine freezing tolerance in North America.

Accurate and real-time monitoring of grapevine freezing tolerance is crucial for the sustainability of the grape industry in cool climate viticultural regions. However, on-site data are limited due to the complexity of measurement. Current prediction models underperform under diverse climate conditions, which limits the large-scale deployment of these methods. We combined grapevine freezing tolerance data from multiple regions in North America and generated a predictive model based on hourly temperature-derived features and cultivar features using AutoGluon, an automated machine learning engine. Feature importance was quantified by AutoGluon and SHAP (SHapley Additive exPlanations) value. The final model was evaluated and compared with previous models for its performance under different climate conditions. The final model achieved an overall 1.36°C root-mean-square error during model testing and outperformed two previous models using three test cultivars at all testing regions. Two feature importance quantification methods identified five shared essential features. Detailed analysis of the features indicates that the model has adequately extracted some biological mechanisms during training. The final model, named NYUS.2, was deployed along with two previous models as an R shiny-based application in the 2022-23 dormancy season, enabling large-scale and real-time simulation of grapevine freezing tolerance in North America for the first time.

Electrical stimulation promoting the angiogenesis in diabetic rat perforator flap through attenuating oxidative stress-mediated inflammation and apoptosis.

Background: Skin flap transplantation is one of the effective methods to treat the diabetes-related foot ulceration, but the intrinsic damage to vessels in diabetes mellitus (DM) leads to the necrosis of skin flaps. Therefore, the discovery of a non-invasive and effective approach for promoting the survival of flaps is of the utmost importance. Electrical stimulation (ES) promotes angiogenesis and increases the proliferation, migration, and elongation of endothelial cells, thus being a potential effective method to improve flap survival. Objective: The purpose of this study was to elucidate the mechanism used by ES to effectively restore the impaired function of endothelial cells caused by diabetes. Methods: A total of 79 adult male Sprague-Dawley rats were used in this study. Gene and protein expression was assessed by PCR and western blotting, respectively. Immunohistochemistry and hematoxylin-eosin staining were performed to evaluate the morphology and density of the microvessels in the flap. Results: The optimal duration for preconditioning the flap with ES was 7 days. The flap survival area percentage and microvessels density in the DMES group were markedly increased compared to the DM group. VEGF, MMP2, and MMP9 protein expression was significantly upregulated. ROS intensity was significantly decreased and GSH concentration was increased. The expression of IL-1ß, MCP­1, cleaved caspase-3, and Bax were downregulated in the DMES group, while TGF-ß expression was upregulated. Conclusions: ES improves the angiogenesis in diabetic ischemic skin flaps by attenuating oxidative stress-mediated inflammation and apoptosis, eventually increasing their viability.

Icariin promotes osteogenic differentiation of human bone marrow mesenchymal stem cells by regulating USP47/SIRT1/Wnt/ß-catenin.

Icariin has been shown to promote osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs). However, the underlying molecular mechanism by which Icariin regulates osteogenic differentiation needs to be further revealed. The viability of BMSCs was assessed by cell counting kit 8 assay. BMSC osteogenic differentiation ability was evaluated by detecting alkaline phosphatase activity and performing alizarin red S staining. The protein levels of osteogenic differentiation-related markers, sirtuin 1 (SIRT1), ubiquitin-specific protease 47 (USP47), and Wnt/ß-catenin-related markers were determined using western blot. SIRT1 mRNA level was measured using quantitative real-time PCR. The regulation of USP47 on SIRT1 was confirmed by ubiquitination detection and co-immunoprecipitation analysis. Icariin could promote BMSC osteogenic differentiation. SIRT1 expression was enhanced by Icariin, and its knockdown suppressed Icariin-induced BMSC osteogenic differentiation. Moreover, deubiquitinating enzyme USP47 could stabilize SIRT1 protein expression. Besides, SIRT1 overexpression reversed the inhibiting effect of USP47 knockdown on BMSC osteogenic differentiation, and USP47 knockdown also restrained Icariin-induced BMSC osteogenic differentiation. Additionally, Icariin enhanced the activity of the Wnt/ß-catenin pathway by upregulating SIRT1. Icariin facilitated BMSC osteogenic differentiation via the USP47/SIRT1/Wnt/ß-catenin pathway.

Deep Learning Detection of Aneurysm Clips for Magnetic Resonance Imaging Safety.

Flagging the presence of metal devices before a head MRI scan is essential to allow appropriate safety checks. There is an unmet need for an automated system which can flag aneurysm clips prior to MRI appointments. We assess the accuracy with which a machine learning model can classify the presence or absence of an aneurysm clip on CT images. A total of 280 CT head scans were collected, 140 with aneurysm clips visible and 140 without. The data were used to retrain a pre-trained image classification neural network to classify CT localizer images. Models were developed using fivefold cross-validation and then tested on a holdout test set. A mean sensitivity of 100% and a mean accuracy of 82% were achieved. Predictions were explained using SHapley Additive exPlanations (SHAP), which highlighted that appropriate regions of interest were informing the models. Models were also trained from scratch to classify three-dimensional CT head scans. These did not exceed the sensitivity of the localizer models. This work illustrates an application of computer vision image classification to enhance current processes and improve patient safety.

Dynamic palmitoylation of STX11 controls injury-induced fatty acid uptake to promote muscle regeneration.

Different types of cells uptake fatty acids in response to different stimuli or physiological conditions; however, little is known about context-specific regulation of fatty acid uptake. Here, we show that muscle injury induces fatty acid uptake in muscle stem cells (MuSCs) to promote their proliferation and muscle regeneration. In humans and mice, fatty acids are mobilized after muscle injury. Through CD36, fatty acids function as both fuels and growth signals to promote MuSC proliferation. Mechanistically, injury triggers the translocation of CD36 in MuSCs, which relies on dynamic palmitoylation of STX11. Palmitoylation facilitates the formation of STX11/SNAP23/VAMP4 SANRE complex, which stimulates the fusion of CD36- and STX11-containing vesicles. Restricting fatty acid supply, blocking fatty acid uptake, or inhibiting STX11 palmitoylation attenuates muscle regeneration in mice. Our studies have identified a critical role of fatty acids in muscle regeneration and shed light on context-specific regulation of fatty acid sensing and uptake.

Gooderoside A from Anoectochilus elatus attenuates acute and chronic pains by inhibiting NO/cGMP and IRAK4/IRAK1/TAK1 signaling pathways.

ETHNOPHARMACOLOGICAL RELEVANCE: Anoectochilus elatus Lindl. was traditionally used for pain treatment and Gooderoside A (GA) was regarded as its principal constituent. AIM OF THE STUDY: To investigate whether GA can be responsible for the antinociceptive activity of A. elatus and explore its underlying mechanism. MATERIALS AND METHODS: Acetic acid-induced abdominal writhing and tail flick tests were employed to evaluate the antinociceptive activity of ethanolic extract of A. elatus (EEA) and GA. Formalin test was used to ascertain the antinociceptive pattern of GA. Entobarbital sodium induced sleep test was adopted to exclude its hypnotic effect, while open-field test was performed to rule out its motor impairment effect. Chronic constriction injury (CCI)-induced neuropathic pain in rats was developed to evaluate its efficacy on neuropathic pain, and BV-2 cells were used to explore the underlying mechanism. RESULTS: EEA and GA, significantly inhibited chemical and thermal nociception. GA suppressed nociception in formalin test in both phase I and II, whereas methylene blue and L-NAME partially reversed its efficacy. GA located inner and slightly blocked sodium channel current, and did not show any hypnotic effect or motor impairment effect. Crucially, GA markedly attenuated chronic neuropathic pain in rats, inhibited the phosphorylation of IRAK4, IRAK1 and TAK1, and suppressed MAPKs pathway in BV-2 cells. CONCLUSION: GA relieved acute and chronic pains in vivo. The mechanism of action involves the blocking of NO/cGMP and IRAK4/IRAK1/TAK1 pathways. These results suggested GA may be a promising candidate for antinociceptive drug development.

SHP-1 Regulates CD8+ T Cell Effector Function but Plays a Subtle Role with SHP-2 in T Cell Exhaustion Due to a Stage-Specific Nonredundant Functional Relay.

SHP-1 (Src homology region 2 domain-containing phosphatase 1) is a well-known negative regulator of T cells, whereas its close homolog SHP-2 is the long-recognized main signaling mediator of the PD-1 inhibitory pathway. However, recent studies have challenged the requirement of SHP-2 in PD-1 signaling, and follow-up studies further questioned the alternative idea that SHP-1 may replace SHP-2 in its absence. In this study, we systematically investigate the role of SHP-1 alone or jointly with SHP-2 in CD8+ T cells in a series of gene knockout mice. We show that although SHP-1 negatively regulates CD8+ T cell effector function during acute lymphocytic choriomeningitis virus (LCMV) infection, it is dispensable for CD8+ T cell exhaustion during chronic LCMV infection. Moreover, in contrast to the mortality of PD-1 knockout mice upon chronic LCMV infection, mice double deficient for SHP-1 and SHP-2 in CD8+ T cells survived without immunopathology. Importantly, CD8+ T cells lacking both phosphatases still differentiate into exhausted cells and respond to PD-1 blockade. Finally, we found that SHP-1 and SHP-2 suppressed effector CD8+ T cell expansion at the early and late stages, respectively, during chronic LCMV infection.

Analysis of risk factors for postoperative deep vein thrombosis after craniotomy and nomogram model construction.

BACKGROUND: Deep vein thrombosis (DVT) of the lower extremity is one of the most common postoperative complications, especially after craniocerebral surgery. DVT may lead to pulmonary embolism, which has a devastating impact on patient prognosis. This study aimed to investigate the incidence and risk factors of DVT in the lower limbs following craniocerebral surgery. AIM: To identify independent risk factors for the development of postoperative DVT and to develop an effective risk prediction model. METHODS: The demographic and clinical data of 283 patients who underwent craniocerebral surgery between December 2021 and December 2022 were retrospectively analyzed. The independent risk factors for lower extremity DVT were identified by univariate and multivariate analyses. A nomogram was created to predict the likelihood of lower extremity DVT in patients who had undergone craniocerebral surgery. The efficacy of the prediction model was determined by receiver operating characteristic curve using the probability of lower extremity DVT for each sample. RESULTS: Among all patients included in the analysis, 47.7% developed lower extremity DVT following craniocerebral surgery. The risk of postoperative DVT was higher in those with a longer operative time, and patients with intraoperative intermittent pneumatic compression were less likely to develop postoperative DVT. CONCLUSION: The incidence of lower extremity DVT following craniocerebral surgery is significant, highlighting the importance of identifying independent risk factors. Interventions such as the use of intermittent pneumatic compression during surgery may prevent the formation of postoperative DVT.

TCI-UNet: transformer-CNN interactive module for medical image segmentation.

Medical image segmentation is a crucial step in developing medical systems, especially for assisting doctors in diagnosing and treating diseases. Currently, UNet has become the preferred network for most medical image segmentation tasks and has achieved tremendous success. However, due to the limitations of convolutional operation mechanisms, its ability to model long-range dependencies between features is limited. With the success of transformers in the computer vision (CV) field, many excellent models that combine transformers with UNet have emerged, but most of them have fixed receptive fields and a single feature extraction method. To address this issue, we propose a transformer-CNN interactive (TCI) feature extraction module and use it to construct TCI-UNet. Specifically, we improve the self-attention mechanism in transformers to enhance the guiding ability of attention maps for computational resource allocation. It can strengthen the network's ability to capture global contextual information from feature maps. Additionally, we introduce local multi-scale information to supplement feature information, allowing the network to focus on important local information while modeling global contextual information. This improves the network's capability to extract feature map information and facilitates effective interaction between global and local information within the transformer, enhancing the representational power of transformers. We conducted a large number of experiments on the LiTS-2017 and ISIC-2018 datasets to verify the effectiveness of our proposed method, with DCIE values of 93.81% and 88.22%, respectively. Through ablation experiments, we proved the effectiveness of the TCI module, and in comparison with other state-of-the-art (SOTA) networks, we demonstrated the superiority of TCI-UNet in accuracy and generalization.

SLC35D3 promotes white adipose tissue browning to ameliorate obesity by NOTCH signaling.

White adipose tissue browning can promote lipid burning to increase energy expenditure and improve adiposity. Here, we show that Slc35d3 expression is significantly lower in adipose tissues of obese mice. While adipocyte-specific Slc35d3 knockin is protected against diet-induced obesity, adipocyte-specific Slc35d3 knockout inhibits white adipose tissue browning and causes decreased energy expenditure and impaired insulin sensitivity in mice. Mechanistically, we confirm that SLC35D3 interacts with the NOTCH1 extracellular domain, which leads to the accumulation of NOTCH1 in the endoplasmic reticulum and thus inhibits the NOTCH1 signaling pathway. In addition, knockdown of Notch1 in mouse inguinal white adipose tissue mediated by orthotopic injection of AAV8-adiponectin-shNotch1 shows considerable improvement in obesity and glucolipid metabolism, which is more pronounced in adipocyte-specific Slc35d3 knockout mice than in knockin mice. Overall, in this study, we reveal that SLC35D3 is involved in obesity via NOTCH1 signaling, and low adipose SLC35D3 expression in obesity might be a therapeutic target for obesity and associated metabolic disorders.

Traditional use, phytochemistry and pharmacology of Viticis Fructus.

Ethnopharmacological relevance: Viticis Fructus (called Manjingzi in China) is the dried ripe fruits of the plant species Vitex trifolia subsp. litoralis Steenis and Vitex trifolia L. in the family Lamiaceae. Viticis Fructus has been used as a traditional Chinese medicine for thousands of years to treat illness such as colds, headache, vertigo, anesthesia, and hyperkinesias. More chemical constituents and medicinal effects have been discovered in Viticis Fructus with the development of modern technology.The aim of the review: This review aims to analyze the research progress of Viticis Fructus from the aspects of botany, ethnopharmacology, phytochemistry, and pharmacological activity, as well as to provide an outlook on the research and use prospects of Viticis Fructus. Material and methods: A comprehensive literature search using online databases such Science Direct, CNKI, Wiley online library, Spring Link, Web of Science, PubMed, Wanfang Data and SCI-Finder. In addition, information was obtained from local and foreign books on ethnobotany and ethnomedicine. Results: The application of Viticis Fructus as a medicine can be traced back to around 480 AD. So far, more than 190 compounds have been isolated from Viticis Fructus, including flavonoids, sterols, cyclic enol ether terpenoids, and diterpenoids. Modern pharmacological studies have shown that the extracts of Viticis Fructus have various pharmacological effects, such as anti-allergic, antioxidant, anti-inflammatory, anti-cancer, and anti-bacterial effects. Conclusion: As a widely used traditional medicine, Viticis Fructus is rich in chemical compositions and has an obvious biological activity. However, the application and pharmacological activity of Viticis Fructus have not been scientifically evaluated or convincing due to poor methodology, unclear results and lack of clinical data. Systematic and comprehensive research evaluations are needed to verify its pharmaceutical activity, clinical therapeutic efficacy and safety. As an important herbal medicine, it should be further explored to facilitate the development of new medicines and treatments for a variety of diseases.

ZFYVE28 mediates insulin resistance by promoting phosphorylated insulin receptor degradation via increasing late endosomes production.

Insulin resistance is associated with many pathological conditions, and an in-depth understanding of the mechanisms involved is necessary to improve insulin sensitivity. Here, we show that ZFYVE28 expression is decreased in insulin-sensitive obese individuals but increased in insulin-resistant individuals. Insulin signaling inhibits ZFYVE28 expression by inhibiting NOTCH1 via the RAS/ERK pathway, whereas ZFYVE28 expression is elevated due to impaired insulin signaling in insulin resistance. While Zfyve28 overexpression impairs insulin sensitivity and causes lipid accumulation, Zfyve28 knockout in mice can significantly improve insulin sensitivity and other indicators associated with insulin resistance. Mechanistically, ZFYVE28 colocalizes with early endosomes via the FYVE domain, which inhibits the generation of recycling endosomes but promotes the conversion of early to late endosomes, ultimately promoting phosphorylated insulin receptor degradation. This effect disappears with deletion of the FYVE domain. Overall, in this study, we reveal that ZFYVE28 is involved in insulin resistance by promoting phosphorylated insulin receptor degradation, and ZFYVE28 may be a potential therapeutic target to improve insulin sensitivity.

Atomically dispersed Fe-O4-C sites as efficient electrocatalysts for electrosynthesis of hydrogen peroxide.

The electrochemical reduction of oxygen via the 2e pathway is an environmentally friendly approach to the electrosynthesis of H2O2. Nevertheless, its sluggish kinetics and limited selectivity hinder its practical application. Herein, single Fe atoms anchored on graphene oxide (SA Fe/GO) with Fe-O4-C sites are developed as an efficient electrocatalyst for the electro-synthesis of H2O2. These Fe-O4-C site active centres could efficiently enhance the activity and selectivity towards 2e electrochemical oxygen reduction in an alkaline environment. The newly-developed SA Fe/GO electrocatalyst demonstrates exceptional electrochemical performance, exhibiting impressive activity with an onset potential of 0.90 and H2O2 production of 0.60 mg cm-2 h-1 at 0.4 V. Remarkably, it achieves a remarkable H2O2 selectivity of over 95.5%.

Physics-Informed Neural Networks for Solving Forward and Inverse Problems in Complex Beam Systems.

This article proposes a new framework using physics-informed neural networks (PINNs) to simulate complex structural systems that consist of single and double beams based on Euler-Bernoulli and Timoshenko theories, where the double beams are connected with a Winkler foundation. In particular, forward and inverse problems for the Euler-Bernoulli and Timoshenko partial differential equations (PDEs) are solved using nondimensional equations with the physics-informed loss function. Higher order complex beam PDEs are efficiently solved for forward problems to compute the transverse displacements and cross-sectional rotations with less than 1e-3 % error. Furthermore, inverse problems are robustly solved to determine the unknown dimensionless model parameters and applied force in the entire space-time domain, even in the case of noisy data. The results suggest that PINNs are a promising strategy for solving problems in engineering structures and machines involving beam systems.

EEG-CDILNet: a lightweight and accurate CNN network using circular dilated convolution for motor imagery classification.

Objective.The combination of the motor imagery (MI) electroencephalography (EEG) signals and deep learning-based methods is an effective way to improve MI classification accuracy. However, deep learning-based methods often need too many trainable parameters. As a result, the trade-off between the network decoding performance and computational cost has always been an important challenge in the MI classification research.Approach.In the present study, we proposed a new end-to-end convolutional neural network (CNN) model called the EEG-circular dilated convolution (CDIL) network, which takes into account both the lightweight model and the classification accuracy. Specifically, the depth-separable convolution was used to reduce the number of network parameters and extract the temporal and spatial features from the EEG signals. CDIL was used to extract the time-varying deep features that were generated in the previous stage. Finally, we combined the features extracted from the two stages and used the global average pooling to further reduce the number of parameters, in order to achieve an accurate MI classification. The performance of the proposed model was verified using three publicly available datasets.Main results.The proposed model achieved an average classification accuracy of 79.63% and 94.53% for the BCIIV2a and HGD four-classification task, respectively, and 87.82% for the BCIIV2b two-classification task. In particular, by comparing the number of parameters, computation and classification accuracy with other lightweight models, it was confirmed that the proposed model achieved a better balance between the decoding performance and computational cost. Furthermore, the structural feasibility of the proposed model was confirmed by ablation experiments and feature visualization.Significance.The results indicated that the proposed CNN model presented high classification accuracy with less computing resources, and can be applied in the MI classification research.