Search

A millimeter-wave radar is widely accepted by the public due to its low susceptibility to interference, such as changes in light, and the protection of personal privacy. With the development of the deep learning theory, the deep learning method has been dominant in the millimeter-wave radar field, which usually uses convolutional neural networks for feature extraction. In recent years, transformer networks have also been highly valued by researchers due to their parallel processing capabilities and long-distance dependency modeling capabilities. However, traditional convolutional neural networks (CNNs) and vision transformers each have their limitations: CNNs usually overlook the global features of images and vision transformers may neglect local image continuity, and both of them may impede gesture recognition performance. In addition, whether CNN or transformer, their implementation is hindered by the scarcity of public radar gesture datasets. To address these limitations, this paper proposes a new recognition method using a local pyramid visual transformer (LPVT) based on millimeter-wave radar. LPVT can capture global and local features in dynamic gesture spectrograms, ultimately improving the recognition ability of gestures. In this paper, we mainly carried out the following two tasks: building the corresponding datasets and executing gesture recognition. First, we constructed a gesture dataset for training. In this stage, we use a 77 GHz radar to collect the echo signals of gestures and preprocess them to build a dataset. Second, we propose the LPVT network specifically designed for gesture recognition tasks. By integrating local sensing into the globally focused transformer, we improve its capacity to capture both global and local features in dynamic gesture spectrograms. The experimental results using the dataset we constructed show that the proposed LPVT network achieved a gesture recognition accuracy of 92.2%, which exceeds the performance of other networks.

Broomcorn millet (Panicum miliaceum L.) is a crop with a good ability to adapt to the environment. Over 8800 accessions have been collected in the national gene bank of China. The huge quantity of germplasms made it difficult for analysis and evaluation. Although a broomcorn millet core collection (CC) comprising 780 accessions was established, the number is still too large for researchers to explore in depth. In this study, the genetic diversity of 634 broomcorn millet accessions from the core collection was analyzed based on SSR markers. A mini-core collection (MC) containing 256 accessions was extracted. The mini-core collection accounted for less than half of the original core collection and only about 2.8% of the total resources but still provided a good representation. In addition, the results of this study validated that Shanxi is the origin of broomcorn millet, and accessions from the South region may contain novel genes. In conclusion, this study provides a comprehensive characterization of the genetic diversities of broomcorn millet core collection in China. Moreover, an MC may aid in reasonably and efficiently selecting materials for broomcorn millet breeding as researchers could screen for aimed genetic characters within a smaller scope.

Atmospheric water vapor is not only a key element of the global hydrological cycle but also the most abundant greenhouse gas. The phase transition and transportation of water vapor are essential for maintaining global energy balance and regulating hydrological processes. However, due to insufficient meteorological observational data, climate research in Africa faces significant limitations despite its substantial contribution to changes in global precipitable water vapor (PWV). In this study, we used MODIS near-infrared (NIR) PWV products and Berkeley temperature data to depict the spatial–temporal variability in PWV across Africa from 2001 to 2020. The results reveal a significant increasing trend in PWV over Africa, with an increase of 0.0158 cm/year. Nearly 99.96% of Africa shows an increase in PWV, with 88.95% of these areas experiencing statistically significant changes, particularly in central regions of Africa. The increase in PWV is more pronounced in high-value months compared to low-value months. The equatorial region of the Congo Basin exhibits higher PWV, which gradually decreases as latitude increases. Despite significant warming (0.0162 °C/year) in Africa, there is no consistent positive correlation between temperature and water vapor. A positive relationship between PWV and temperature is observed in western Africa, while a negative relationship is noted in eastern and southern Africa on an annual scale. Additionally, an increasing trend in precipitation (4.6669 mm/year) is observed, with a significant positive correlation between PWV and precipitation across most of Africa, although this relationship varies by month. These findings provide valuable insights into the comprehension of the hydrothermal variation in Africa amidst climate warming.

Time–frequency parameterization for oscillations in specific frequency bands reflects the dynamic changes in the brain. It is related to cognitive behavior and diseases and has received significant attention in neuroscience. However, many studies do not consider the impact of the aperiodic noise and neural activity, including their time-varying fluctuations. Some studies are limited by the low resolution of the time–frequency spectrum and parameter-solved operation. Therefore, this paper proposes super-resolution time–frequency periodic parameterization of (transient) oscillation (STPPTO). STPPTO obtains a super-resolution time–frequency spectrum with Superlet transform. Then, the time–frequency representation of oscillations is obtained by removing the aperiodic component fitted in a time-resolved way. Finally, the definition of transient events is used to parameterize oscillations. The performance of this method is validated on simulated data and its reliability is demonstrated on magnetoencephalography. We show how it can be used to explore and analyze oscillatory activity under rhythmic stimulation.

Multivariate pattern analysis (MVPA) has played an extensive role in interpreting brain activity, which has been applied in studies with modalities such as functional Magnetic Resonance Imaging (fMRI), Magnetoencephalography (MEG) and Electroencephalography (EEG). The advent of wearable MEG systems based on optically pumped magnetometers (OPMs), i.e., OP-MEG, has broadened the application of bio-magnetism in the realm of neuroscience. Nonetheless, it also raises challenges in temporal decoding analysis due to the unique attributes of OP-MEG itself. The efficacy of decoding performance utilizing multimodal fusion, such as MEG-EEG, also remains to be elucidated. In this regard, we investigated the impact of several factors, such as processing methods, models and modalities, on the decoding outcomes of OP-MEG. Our findings indicate that the number of averaged trials, dimensionality reduction (DR) methods, and the number of cross-validation folds significantly affect the decoding performance of OP-MEG data. Additionally, decoding results vary across modalities and fusion strategy. In contrast, decoder type, resampling frequency, and sliding window length exert marginal effects. Furthermore, we introduced mutual information (MI) to investigate how information loss due to OP-MEG data processing affect decoding accuracy. Our study offers insights for linear decoding research using OP-MEG and expand its application in the fields of cognitive neuroscience.

A class of algorithms based on subspace projection is widely used in the denoising of magnetoencephalography (MEG) signals. Setting the dimension of the interference (external) subspace matrix of these algorithms is the key to balancing the denoising effect and the degree of signal distortion. However, most current methods for estimating the dimension threshold rely on experience, such as observing the signal waveforms and spectrum, which may render the results too subjective and lacking in quantitative accuracy. Therefore, this study proposes a method to automatically estimate a suitable threshold. Time–frequency transformations are performed on the evoked state data to obtain the neural signal of interest and the noise signal in a specific time–frequency band, which are then used to construct the objective function describing the degree of noise suppression and signal distortion. The optimal value of the threshold in the selected range is obtained using the weighted-sum method. Our method was tested on two classical subspace projection algorithms using simulation and two sensory stimulation experiments. The thresholds estimated by the proposed method enabled the algorithms to achieve the best waveform recovery and source location error. Therefore, the threshold selected in this method enables subspace projection algorithms to achieve the best balance between noise removal and neural signal preservation in subsequent MEG analyses.

Synthetic aperture radar (SAR) images have special physical scattering characteristics owing to their unique imaging mechanism. Traditional deep learning algorithms usually extract features from real-valued SAR images in a purely data-driven manner, which may ignore some important physical scattering characteristics and sacrifice some useful target information in SAR images. This undoubtedly limits the improvement in performance for SAR target recognition. To take full advantage of the physical information contained in SAR images, a complex-valued network guided with sub-aperture decomposition (CGS-Net) for SAR target recognition is proposed. According to the fact that different targets have different physical scattering characteristics at different angles, the sub-aperture decomposition is used to improve accuracy with a multi-task learning strategy. Specifically, the proposed method includes main and auxiliary tasks, which can improve the performance of the main task by learning and sharing useful information from the auxiliary task. Here, the main task is the target recognition task, and the auxiliary task is the target reconstruction task. In addition, a complex-valued network is used to extract the features from the original complex-valued SAR images, which effectively utilizes the amplitude and phase information in SAR images. The experimental results obtained using the MSTAR dataset illustrate that the proposed CGS-Net achieved an accuracy of 99.59% (without transfer learning or data augmentation) for the ten-classes targets, which is superior to the other popular deep learning methods. Moreover, the proposed method has a lightweight network structure, which is suitable for SAR target recognition tasks because SAR images usually lack a large number of labeled data. Here, the experimental results obtained using the small dataset further demonstrate the excellent performance of the proposed CGS-Net.

Glass Fiber reinforced polymers (GFRPs) are widely used and play an important role in modern society. The multilayer structure of GFRPs can lead to delamination defects during production and service, which can have a significant impact on the integrity and safety of the equipment. Therefore, it is important to monitor these delamination defects during equipment service in order to evaluate their effects on equipment performance and lifespan. Microwave imaging testing, with its high sensitivity and noncontact nature, shows promise as a potential method for detecting delamination defects in GFRPs. However, there is currently limited research on the quantitative characterization of defect images in this field. In order to achieve visual quantitative nondestructive testing (NDT), we propose a 2D-imaging visualization and quantitative characterization method for delamination defects in GFRP, and realize the combination of visual detection and quantitative detection. We built a microwave testing experimental system to verify the effectiveness of the proposed method. The results of the experiment indicate the effectiveness and innovation of the method, which can effectively detect all delamination defects of 0.5 mm thickness inside GFRP with high accuracy, the signal-to-background ratio (SBR) of 2D imaging can reach 4.41 dB, the quantitative error of position is within 0.5 mm, and the relative error of area is within 11%.

Glass fiber-reinforced polymer (GFRP) is widely used in engineering fields involving aerospace, energy, transportation, etc. If internal buried defects occur due to hostile environments during fabrication and practical service, the structural integrity and safety of GFRP structures would be severely undermined. Therefore, it is indispensable to carry out effective quantitative nondestructive testing (NDT) of internal defects buried within GFRP structures. Along with the development of composite materials, microwave NDT is promising in non-intrusive inspection of defects in GFRPs. In this paper, quantitative screening of the subsurface impact damage and air void in a unidirectional GFRP via microwave reflectometry was intensively investigated. The influence of the microwave polarization direction with respect to the GFRP fiber direction on the reflection coefficient was investigated by using the equivalent relative permittivity calculated with theoretical analysis. Following this, a microwave NDT system was built up for further investigation regarding the imaging and quantitative evaluation of buried defects in GFRPs. A direct-wave suppression method based on singular-value decomposition was proposed to obtain high-quality defect images. The defect in-plane area was subsequently assessed via a proposed defect-edge identification method. The simulation and experimental results revealed that (1) the testing sensitivity to buried defects was the highest when the electric-field polarization direction is parallel to the GFRP fiber direction; and (2) the averaged evaluation accuracy regarding the in-plane area of the buried defect reached approximately 90% by applying the microwave reflectometry together with the proposed processing methods.

N6-methyladenosine (m6A) plays a significant role as an epigenetic mechanism, which is involved in various cancers’ progress via regulating mRNA modification. As a crucial m6A “reader”, YTHDF1 is able to alter m6A-modified mRNA and promote the protein translation process in multiple cancers. However, the role of YTHDF1 in lung cancer has not been fully investigated. This study focuses on elucidating the function of YTHDF1 in the development of lung cancer and its underlying mechanism. We demonstrated that YTHDF1 was highly expressed in lung carcinoma progression; then, the loss of function experiments in lung cell lines confirmed that knockdown of YTHDF1 suppressed cell proliferation, migration and invasion and induced ferroptosis of lung cancer cells. Further functional assays showed that ferritin (FTH) was identified as the key target of YTHDF1 in lung cancer cells. Furthermore, the overexpression of ferritin in YTHDF1-depleted cells partially restored lung cancer cell suppression. Collectively, our data suggested that the upregulation of YTHDF1 promotes lung cancer carcinogenesis by accelerating ferritin translation in an m6A-dependent manner. We hope that our findings may provide a new target for lung cancer diagnosis and treatment.

Based on the emission factor method and LMDI-LEAP model, this paper systematically studies the current situation, influencing factors and changing trend of carbon emissions from rural buildings in a typical village located in southern China. The results showed that (1) the per capita carbon emissions generated by the energy consumption of rural buildings is 2.58 tCO2/a. Carbon emissions from electricity consumption in buildings account for about 96.07%; (2) the per capita building area, building area energy intensity, population size, population structure and carbon emission coefficient affect rural building carbon emissions, with contribution rates of 70.13%, 31.27%, 0.61%, −1.21% and −0.80%, respectively; (3) from 2021 to 2060, the carbon emissions of rural buildings are expected to increase first and then decrease. In 2021, the base year, carbon emissions from buildings were 2755.49 tCO2. The carbon emissions will peak at 5275.5 tCO2. Measures such as controlling the scale of buildings and improving the utilization rate of clean energy can effectively reduce carbon emissions, in which case the peak can be reduced to 4830.06 tCO2. Finally, the countermeasures and suggestions about rural building energy saving and emission reduction are proposed, including improving the construction management, raising energy efficiency standards in buildings, increasing the proportion of clean energy and raising residents’ awareness of energy conservation.

Diabetic retinopathy (DR), as the leading cause of vision loss in the working-age population, exhibits unique metabolite profiles in human plasma and vitreous. However, those in retina are not fully understood. Here, we utilized liquid and gas chromatography–tandem mass spectrometry technology to explore metabolite characteristics of streptozotocin (STZ)-induced diabetic mice retina. A total of 145 metabolites differed significantly in diabetic retinas compared with controls. These metabolites are mainly enriched in the Warburg effect, and valine, leucine and isoleucine degradation pathways. To further identify underlying regulators, RNA sequencing was performed to integrate metabolic enzyme alterations with metabolomics in STZ-induced diabetic retina. Retinol metabolism and tryptophan metabolism are the shared pathways enriched by metabolome and transcriptome. Additionally, transcriptomic analysis identified 71 differentially expressed enzyme-related genes including Hk2, Slc7a5, Aldh1a3 and Tph integrated with altered metabolic pathways. In addition, single nucleotide polymorphisms within 6 out of 71 genes are associated with increased diabetes risk. This study lays the foundation for mechanism research and the therapeutic target development of DR.

It is of great significance for air pollution control and personnel safety guarantee to master the release characteristics of harmful gases in the process of Limnoperna fortunei corruption. In view of the lack of research on the environmental pollution caused by the corruption of Limnoperna fortunei, a model experiment was designed to study the three harmful gases of NH3, H2S, and CH4 in the putrid process of Limnoperna fortunei by considering the density of Limnoperna fortunei and the time of leaving water. The results show that: (1) The recognition and processing of outliers based on wavelet decomposition and K-means algorithm can effectively reduce the standard deviation and coefficient of variation of the data set and improve the accuracy of the data set. (2) The variation of NH3 and H2S gas concentrations with the time of water separation satisfies polynomial linear regression (R2 > 99%). (3) At a density of 0.5–7.0 × 104 mussels/m2, the highest concentration of NH3 reached 47.9777–307.9454 mg/m3 with the increase in the density of Limnoperna fortunei and the extension of the time away from water, far exceeding the occupational exposure limit of NH3 of 30 mg/m3, potentially threatening human health and safety. The highest detection value of H2S concentration is 0.1909–5.0946 mg/m3, and the highest detection concentration of CH4 is 0.02%, both of which can be ignored.