OD-YOLO: Robust Small Object Detection Model in Remote Sensing Image with a Novel Multi-Scale Feature Fusion.

As remote sensing technology has advanced, the use of satellites and similar technologies has become increasingly prevalent in daily life. Now, it plays a crucial role in hydrology, agriculture, and geography. Nevertheless, because of the distinct qualities of remote sensing, including expansive scenes and small, densely packed targets, there are many challenges in detecting remote sensing objects. Those challenges lead to insufficient accuracy in remote sensing object detection. Consequently, developing a new model is essential to enhance the identification capabilities for objects in remote sensing imagery. To solve these constraints, we have designed the OD-YOLO approach that uses multi-scale feature fusion to improve the performance of the YOLOv8n model in small target detection. Firstly, traditional convolutions have poor recognition capabilities for certain geometric shapes. Therefore, in this paper, we introduce the Detection Refinement Module (DRmodule) into the backbone architecture. This module utilizes Deformable Convolutional Networks and the Hybrid Attention Transformer to strengthen the model's capability for feature extraction from geometric shapes and blurred objects effectively. Meanwhile, based on the Feature Pyramid Network of YOLO, at the head of the model framework, this paper enhances the detection capability by introducing a Dynamic Head to strengthen the fusion of different scales features in the feature pyramid. Additionally, to address the issue of detecting small objects in remote sensing images, this paper specifically designs the OIoU loss function to finely describe the difference between the detection box and the true box, further enhancing model performance. Experiments on the VisDrone dataset show that OD-YOLO surpasses the compared models by at least 5.2% in mAP50 and 4.4% in mAP75, and experiments on the Foggy Cityscapes dataset demonstrated that OD-YOLO improved mAP by 6.5%, demonstrating outstanding results in tasks related to remote sensing images and adverse weather object detection. This work not only advances the research in remote sensing image analysis, but also provides effective technical support for the practical deployment of future remote sensing applications.

TW-YOLO: An Innovative Blood Cell Detection Model Based on Multi-Scale Feature Fusion.

As deep learning technology has progressed, automated medical image analysis is becoming ever more crucial in clinical diagnosis. However, due to the diversity and complexity of blood cell images, traditional models still exhibit deficiencies in blood cell detection. To address blood cell detection, we developed the TW-YOLO approach, leveraging multi-scale feature fusion techniques. Firstly, traditional CNN (Convolutional Neural Network) convolution has poor recognition capabilities for certain blood cell features, so the RFAConv (Receptive Field Attention Convolution) module was incorporated into the backbone of the model to enhance its capacity to extract geometric characteristics from blood cells. At the same time, utilizing the feature pyramid architecture of YOLO (You Only Look Once), we enhanced the fusion of features at different scales by incorporating the CBAM (Convolutional Block Attention Module) in the detection head and the EMA (Efficient Multi-Scale Attention) module in the neck, thereby improving the recognition ability of blood cells. Additionally, to meet the specific needs of blood cell detection, we designed the PGI-Ghost (Programmable Gradient Information-Ghost) strategy to finely describe the gradient flow throughout the process of extracting features, further improving the model's effectiveness. Experiments on blood cell detection datasets such as BloodCell-Detection-Dataset (BCD) reveal that TW-YOLO outperforms other models by 2%, demonstrating excellent performance in the task of blood cell detection. In addition to advancing blood cell image analysis research, this work offers strong technical support for future automated medical diagnostics.

Uncovering the spatio-temporal dynamics of crop-specific nutrient budgets in China.

Nitrogen (N) and phosphorus (P) are two critical nutrients for agroecosystems. In meeting food demands, human use of both nutrients has crossed planetary boundaries for sustainability. Further, there has been a dramatic shift in their relative inputs and outputs, which may generate strong N:P imbalances. Despite enormous efforts on agronomic N and P budgets, the spatio-temporal characteristics of different crop types in using nutrients are unknown as are patterns in the stoichiometric coupling of these nutrients. Thus, we analyzed the annual crop-specific N and P budgets and their stoichiometric relations for producing ten major crops at the provincial level of China during 2004-2018. Results show that, China has generally witnessed excessive N and P input over the past 15 years, with the N balance remaining stable while the P balance increasing by more than 170%, thus resulting in a decline in the N:P mass ratios from 10.9 in 2004 to 3.8 in 2018. Crop-aggregated nutrient use efficiency (NUE) of N has increased by 10% in these years while most crops have shown a decreasing trend of this indicator for P, which reduced NUE of P from 75% to 61% during this period. At the provincial level, the nutrient fluxes of Beijing and Shanghai have obviously declined, while the nutrient fluxes of provinces such as Xinjiang and Inner Mongolia have increased significantly. Although N management has made progress, P management should be further explored in the future due to eutrophication concerns. More importantly, N and P management strategies for sustainable agriculture in China should take account of not only the absolute nutrient use, but also their stoichiometric balance for different crops in different locations.

Rasterizing CO2 emissions and characterizing their trends via an enhanced population-light index at multiple scales in China during 2013-2019.

Climate change caused by CO2 emissions (CE) has received widespread global concerns. Obtaining precision CE data is necessary for achieving carbon peak and carbon neutrality. Significant deficiencies of existing CE datasets such as coarse spatial resolution and low precision can hardly meet the actual requirements. An enhanced population-light index (RPNTL) was developed in this study, which integrates the Nighttime Light Digital Number (DN) Value from the National Polar-orbiting Partnership (NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) and population density to improve CE estimation accuracy. The CE from the Carbon Emission Accounts & Datasets (CEADS) was divided into three sectors, namely urban, industrial, and rural, to differentiate the heterogeneity of CE in each sector. The ordinary least square (OLS), geographically weighted regression (GWR), temporally weighted regression (TWR), and geographically and temporally weighted regression (GTWR) models were employed to establish the quantitative relationship between RPNTL and CE for each sector. The optimal model was determined through model comparison and precision evaluation and was utilized to rasterize CE for urban, industrial, and rural areas. Additionally, hot spot analysis, trend analysis, and standard deviation ellipses were introduced to demonstrate the spatiotemporal dynamic characteristics of CE at multiple scales. The performance of the GTWR outperformed other methods in estimating CE. The enhanced RPNTL demonstrated a higher coefficient of determination (R2 = 0.95) than the NTL (R2 = 0.92) in predicting CE, particularly in rural regions where the R2 value increased from 0.76 to 0.81. From 2013 to 2019, high CE was observed in eastern and northern China, while a decreasing trend was detected in northeastern China and Chengdu-Chongqing. Conversely, the Yangtze River Delta, Pearl River Delta, Fenwei Plain, and Henan Province showed an increasing trend. The center of gravity for industrial and rural CE is shifting towards western regions, whereas that for urban CE is moving northward. This study provides valuable insights for decision-making on CE control.

Study on the spatiotemporal dynamic of ground-level ozone concentrations on multiple scales across China during the blue sky protection campaign.

Surface ozone (O3), one of the harmful air pollutants, generated significantly negative effects on human health and plants. Existing O3 datasets with coarse spatiotemporal resolution and limited coverage, and the uncertainties of O3 influential factors seriously restrain related epidemiology and air pollution studies. To tackle above issues, we proposed a novel scheme to estimate daily O3 concentrations on a fine grid scale (1 km × 1 km) from 2018 to 2020 across China based on machine learning methods using hourly observed ground-level pollutant concentrations data, meteorological data, satellite data, and auxiliary data including digital elevation model (DEM), land use data (LUD), normalized difference vegetation index (NDVI), population (POP), and nighttime light images (NTL), and to identify the difference of influential factors of O3 on diverse urbanization and topography conditions. Some findings were achieved. The correlation coefficients (R2) between O3 concentrations and surface net solar radiation (SNSR), boundary layer height (BLH), 2 m temperature (T2M), 10 m v-component (MVW), and NDVI were 0.80, 0.40, 0.35, 0.30, and 0.20, respectively. The random forest (RF) demonstrated the highest validation R2 (0.86) and lowest validation RMSE (13.74 µg/m3) in estimating O3 concentrations, followed by support vector machine (SVM) (R2 = 0.75, RMSE = 18.39 µg/m3), backpropagation neural network (BP) (R2 = 0.74, RMSE = 19.26 µg/m3), and multiple linear regression (MLR) (R2 = 0.52, RMSE = 25.99 µg/m3). Our China High-Resolution O3 Dataset (CHROD) exhibited an acceptable accuracy at different spatial-temporal scales. Additionally, O3 concentrations showed decreasing trend and represented obviously spatiotemporal heterogeneity across China from 2018 to 2020. Overall, O3 was mainly affected by human activities in higher urbanization regions, while O3 was mainly controlled by meteorological factors, vegetation coverage, and elevation in lower urbanization regions. The scheme of this study is useful and valuable in understanding the mechanism of O3 formation and improving the quality of the O3 dataset.

The influence of socioeconomic and environmental determinants on acute myocardial infarction (AMI) mortality from the spatial epidemiological perspective.

Plenty of epidemiological approaches have been explored to detect the effects of environmental and socioeconomic factors on acute myocardial infarction (AMI) mortality. Whereas, identifying the influence of potential affecting factors on AMI mortality based on a spatial epidemiological perspective was strongly desired. Moreover, the interaction effects of two potential factors on the diseases were always neglected previously. Here, the Geodetector and geographically & temporally weighted regression model (GTWR) combined with multi-source spatiotemporal datasets were introduced to quantitatively determine the relationship between AMI mortality and potential influencing factors across Xi'an during 2014-2016. Besides, Moran's I was adopted to diagnose the spatial autocorrelation of AMI mortality. Some findings were achieved. The number of AMI mortality cases increased from 5075 in 2014 to 6774 in 2016. Air pollutants, meteorological factors, economic status, and topography factors exhibited a significant effect on AMI mortality. The AMI mortality demonstrated an obvious spatial autocorrelation feature during 2014-2016. POP and PE represented the most obvious impact on AMI mortality, respectively. Moreover, the interaction of any two factors was larger than that of the single factor on AMI mortality, and the factors with the strongest interaction vary according to lag groups and ages. The effects of factors on AMI mortality were POP (- 628.925) > PE (140.102) > RD (79.145) > O3 (- 58.438) > E_NH3 (42.370) for male, and POP (- 751.206) > RD (132.935) > E_NH3 (58.758) > PE (- 45.434) > O3 (- 21.256) for female, respectively. This work reminds the local government to continuously control air pollution, strengthen urban planning, and improve the health care of the rural areas for alleviating AMI mortality. Meanwhile, the scheme of the current study supplies a scientific reference for examining the effects of potential impact factors on related diseases using the spatial epidemiological perspective.

Identifying the spatiotemporal dynamic of PM2.5 concentrations at multiple scales using geographically and temporally weighted regression model across China during 2015-2018.

Fine particulate matter (PM2.5) is closely related to the air quality and public health. Numerous models have been introduced to simulate the PM2.5 concentrations at large scale based on remote sensing and auxiliary data. However, the data precision provided by these models are inadequate for epidemiology and pollutant exposure studies at medium or small scale. The present study aims to calibrate PM2.5 concentrations at 1 km resolution scale across China during 2015-2018 based on monitoring station data and auxiliary data using a novel geographically and temporally weighted regression model (GTWR). The cross-validation (CV) method and the geographically weighted regression (GWR) model are conducted for validation and cross-comparison. Additionally, the spatial autocorrelation and slope analysis methods are implemented to detect the spatiotemporal dynamic of PM2.5 concentrations. A sample-based CV of the GTWR model demonstrates an acceptable precision with a coefficient of determination equal to 0.67, a root-mean-square error of 10.32 µg/m3, and a mean prediction error of-6.56 µg/m3. This result proves that the GTWR model can simulate PM2.5 concentrations at a higher spatial resolution and accuracy across China than some previous models. Besides, the heterogeneity and spatiotemporal dynamic of PM2.5 concentrations are obvious, that is, the High-High (H-H) agglomeration areas with strong haze pollution were mainly concentrated in Beijing-Tianjin-Hebei (BTH), Yangtze River Delta (YRD), Chengdu-Chongqing (CY), and Guanzhong Plain (GZP). In addition, the PM2.5 concentrations are undergoing a decreasing trend in most of the study area, and the decrease in the BTH is dramatic. The results of the present study are helpful for calibrating and detecting the spatiotemporal dynamic of PM2.5 concentrations and useful for the government to make decisions about decreasing haze pollution in urban agglomeration scale.

Estimating PM2.5 concentrations via random forest method using satellite, auxiliary, and ground-level station dataset at multiple temporal scales across China in 2017.

Fine particulate matter with aerodynamic diameters less than 2.5 µm (PM2.5) poses adverse impacts on public health and the environment. It is still a great challenge to estimate high-resolution PM2.5 concentrations at moderate scales. The current study calibrated PM2.5 concentrations at a 1 km resolution scale using ground-level monitoring data, Aerosol Optical Depth (AOD), meteorological data, and auxiliary data via Random Forest (RF) model across China in 2017. The three ten-folded cross-validations (CV) methods including sample-based, time-based, and spatial-based validation combined with Coefficient Square (R2), Root-Mean-Square Error (RMSE), and Mean Predictive Error (MPE) have been used for validation at different temporal scales in terms of daily, monthly, heating seasonal, and non-heating seasonal. Finally, the distribution map of PM2.5 concentrations was illustrated based on the RF model. Some findings were achieved. The RF model performed well, with a relatively high sample-based cross-validation R2 of 0.74, a low RMSE of 16.29 µg × m-3, and a small MPE of -0.282 µg × m-3. Meanwhile, the performance of the RF model in inferring the PM2.5 concentrations was well at urban scales except for Chengyu (CY). North China, the CY urban agglomeration, and the northwest of China exhibited relatively high PM2.5 pollution features, especially in the heating season. The robustness of the RF model in the present study outperformed most statistical regression models for calibrating PM2.5 concentrations. The outcomes can supply an up-to-date scientific dataset for epidemiological and air pollutants exposure risk studies across China.

Retrieving zinc concentrations in topsoil with reflectance spectroscopy at Opencast Coal Mine sites.

Heavy metals contaminations in mining areas aroused wide concerns globally. Efficient evaluation of its pollution status is a basis for further soil reclamation. Visible and near-infrared reflectance (Vis-NIR) spectroscopy has been diffusely used for retrieving heavy metals concentrations. However, the reliability and feasibility of calibrated models were still doubtful. The present study estimated zinc (Zn) concentrations via the random forest (RF) and partial least squares regression (PLSR) using ground in-situ Zn concentrations as well as soil spectral reflectance at an Opencast Coal Mine of Ordos, China in February 2020. The coefficient of determination (R2), root mean square error (RMSE), mean absolute error (MAE), and the ratio of performance to deviation (RPD) were selected to assess the robustness of the methods in estimating Zn contents. Moreover, the characteristic bands were chosen by Pearson correlation analysis and Boruta Algorithm. Finally, the comparison between RF and PLSR combined with eight spectral reflectance transformation methods was conducted for four concentration groups to determine the optimal model. The results indicated that: (1) Zn contents represented a skewed distribution (coefficient of variation (CV) = 33%); (2) the spectral reflectance tended to decrease with the increase of Zn contents during 580-1850 nm based on Savitzky-Golay smoothing (SG); (3) the continuous wavelet transform (CWT) demonstrated higher effectiveness than other spectral reflectance transformation methods in enhancing spectral responses, the R2 between Zn contents and the soil spectral reflectance achieved the highest (R2 = 0.71) by using CWT; (4) the RF combined with CWT exhibited the best performance than other methods in the current study (R2 = 0.97, RPD = 3.39, RMSE = 1.05 mg kg-1, MAE = 0.79 mg kg-1). The current study supplied a scientific scheme and theoretical support for predicting heavy metals concentrations via the Vis-NIR spectral method in possible contaminated areas such as coal mines and metallic mineral deposit areas.

Determining the effects of socioeconomic and environmental determinants on chronic obstructive pulmonary disease (COPD) mortality using geographically and temporally weighted regression model across Xi'an during 2014-2016.

Numerous methods have been implemented to evaluate the relationship between environmental factors and respiratory mortality. However, the previous epidemiological studies seldom considered the spatial and temporal variation of the independent variables. The present study aims to detect the relations between respiratory mortality and related affecting factors across Xi'an during 2014-2016 based on a novel geographically and temporally weighted regression model (GTWR). Meanwhile, the ordinary least square (OLS) and the geographically weighted regression (GWR) models were developed for cross-comparison. Additionally, the spatial autocorrelation and Hot Spot analysis methods were conducted to detect the spatiotemporal dynamic of respiratory mortality. Some important outcomes were obtained. Socioeconomic and environmental determinants represented significant effects on respiratory diseases. The respiratory mortality exhibited an obvious spatial correlation feature, and the respiratory diseases tend to occur in winter and rural areas of the study area. The GTWR model outperformed OLS and GWR for determining the relations between respiratory mortality and socioeconomic as well as environmental determinants. The influence degree of anthropic factors on COPD mortality was higher than natural factors, and the effects of independent variables on COPD varied timely and locally. The results can supply a scientific basis for respiratory disease controlling and health facilities planning.

Ecological risk evaluation and source apportionment of heavy metals in park playgrounds: a case study in Xi'an, Shaanxi Province, a northwest city of China.

Park playgrounds recently are suffering serious heavy metals contamination in China. It is urgent to assess the ecological risk and identify the sources for heavy metals. A total of 111 topsoil samples were collected from four park playgrounds in Xi'an, and the X-ray fluorescence (XRF) instrument was used to measure the concentrations of heavy metals including chromium(Cr), nickel (Ni), copper (Cu), zinc (Zn), arsenic (As), lead (Pb), manganese (Mn), and cobalt (Co), respectively. Ecological risk ([Formula: see text]) and potential ecological risk index (RI) were introduced to determine the pollution level and ecological risk, and the absolute principal component score-multiple linear regression (APCS-MLR) model was implemented to identify the sources for heavy metals. The main results were as follows. (1) Except As, the mean concentrations of measured heavy metals of four park playgrounds surpassed the soil background values of Shaanxi Province. (2) In each park playground, the [Formula: see text] was below a "low" risk level ([Formula: see text]=10) for Cr, Ni, Zn, As, and Mn; Cu was between a "moderate" and "considerable" risk level; Pb was between a "low" and "moderate" risk level; and [Formula: see text] was between a "considerable" and "high" risk level for Co. Besides, the RI index was on a "high" risk level (120 < RI < 240) with an obvious spatial distinction. (3) The anthropogenic factors were the main sources for heavy metals, and mixed sources and natural sources were considered as the minor sources for metals. (4) The sources contributions for Co had obvious spatial heterogeneity in each park situated in four different urban planning districts.

Expression of osteoprotegerin and receptor activator of nuclear factor κB ligand in root resorption induced by heavy force in rats.

OBJECTIVE: The aim of this study was to investigate the expression patterns of osteoprotegerin (OPG) and the receptor activator of nuclear factor κB ligand (RANKL) in root resorption during orthodontic tooth movement. MATERIAL AND METHODS: Forty 12-week-old male SD rats were used with the right maxillary side as the experimental group and the left maxillary side as the control group. After 1 N (100g) force was loaded on the right maxillary first molar, the rats were sacrificed on days 0, 1, 4, 8, and 12. Mesial root resorption of the first molar, the number of odontoclasts and osteoclasts, and OPG and RANKL mRNA expression were determined by hematoxylin-eosin and scanning electron microscopy, tartrate-resistant acid phosphate staining, and in situ hybridization, respectively. RESULTS: Serious root resorption was apparent on the pressure side of the mesial root of the right maxillary first molar on days 8 and 12. The number of odontoclasts in the cementum lacuna was elevated on days 8 and 12. OPG expression rose significantly on the tensile side, while RANKL expression increased on the pressure side. The mRNA level of RANKL was significantly elevated on days 4, 8, and 12. Moreover, the RANKL/OPG mRNA ratio was increased on the pressure side, but decreased on the tensile side. CONCLUSION: Changes in the expression of RANKL mRNA and the RANKL/OPG mRNA ratio are accompanied by a parallel alteration in the number of odontoclasts and tooth resorption, suggesting crucial involvement of RANKL and OPG in tooth resorption.