Catalogue Search | MBRL
Are you sure you want to remove the book from the shelf?
{{itemTitle}}
1,187
result(s) for
"Li, Ziming"
Sort by:
Analysis of Flood Carrying Capacity Change of Small and Medium-sized Rivers in the Upper Reaches of the Zhanghe River in Southern Shanxi
Using hydrological data and historical flood survey data of the design rivers, the cross-sections of different reaches are measured, and the hydraulic factors of each cross-section are calculated. HEC-RAS (5.0.7) software is used to analyze and calculate the flood water surface lines of different reaches. Taking the cross-sections measured furthest upstream as the starting points and flood peak discharges of 2% flood frequency and the corresponding water levels as the initial and boundary conditions, the design water surface lines of the design reaches are calculated, And recheck calculation of embankment top elevations is performed. And the results show that: (1) the main channels of the rivers shrink continuously due to the short flood discharge time and the influence of farming and other human activities; (2) the average longitudinal slopes of the Cuiguo Village and Du Village reaches of the Jimingshui River are 9.8 ‰ and 7.2 ‰ respectively, and the average longitudinal slope of the Xixiawang Village reach of the Qinjiagou River is 8.6 ‰; (3) the roughness values of different reaches are analyzed, and the calculated design flood water levels of the Cuiguo Village and Du Village reaches of the Jimingshui River and the Xixiawang Village reach of the Qinjiagou River when a great flood with P=2% occurs are 966.6 m, 938.05 m and 931.16 m, respectively, which are all lower than the embankment top elevations of the left and right banks by more than 2 meters, indicating that the flood in these reaches will not overflow and can pass safely.
Journal Article
A CNN-LSTM Car-Following Model Considering Generalization Ability
To explore the potential relationship between the leading vehicle and the following vehicle during car-following, we proposed a novel car-following model combining a convolutional neural network (CNN) with a long short-term memory (LSTM) network. Firstly, 400 car-following periods were extracted from the natural driving database and the OpenACC car-following experiment database. Then, we developed a CNN-LSTM car-following model, and the CNN is employed to analyze the potential relationship between the vehicle’s dynamic parameters and to extract the features of car-following behavior to generate the feature vector. The LSTM network is adopted to save the feature vector and predict the speed of the following vehicle. Finally, the CNN-LSTM model is trained and tested with the extracted car-following trajectories data and compared with the classical car-following models (LSTM model, intelligent driver model). The results show that the accuracy and the ability to learn the heterogeneity of the proposed model are better than the other two. Furthermore, the CNN-LSTM model can accurately reproduce the hysteresis phenomenon of congested traffic flow and apply to heterogeneous traffic flow mixed with adaptive cruise control vehicles on the freeway, which indicates that it has strong generalization ability.
Journal Article
A Deep Learning-Based Framework for Automated Extraction of Building Footprint Polygons from Very High-Resolution Aerial Imagery
Accurate building footprint polygons provide essential data for a wide range of urban applications. While deep learning models have been proposed to extract pixel-based building areas from remote sensing imagery, the direct vectorization of pixel-based building maps often leads to building footprint polygons with irregular shapes that are inconsistent with real building boundaries, making it difficult to use them in geospatial analysis. In this study, we proposed a novel deep learning-based framework for automated extraction of building footprint polygons (DLEBFP) from very high-resolution aerial imagery by combining deep learning models for different tasks. Our approach uses the U-Net, Cascade R-CNN, and Cascade CNN deep learning models to obtain building segmentation maps, building bounding boxes, and building corners, respectively, from very high-resolution remote sensing images. We used Delaunay triangulation to construct building footprint polygons based on the detected building corners with the constraints of building bounding boxes and building segmentation maps. Experiments on the Wuhan University building dataset and ISPRS Vaihingen dataset indicate that DLEBFP can perform well in extracting high-quality building footprint polygons. Compared with the other semantic segmentation models and the vector map generalization method, DLEBFP is able to achieve comparable mapping accuracies with semantic segmentation models on a pixel basis and generate building footprint polygons with concise edges and vertices with regular shapes that are close to the reference data. The promising performance indicates that our method has the potential to extract accurate building footprint polygons from remote sensing images for applications in geospatial analysis.
Journal Article
FGFR1-ERK1/2-SOX2 axis promotes cell proliferation, epithelial–mesenchymal transition, and metastasis in FGFR1-amplified lung cancer
Epithelial–mesenchymal transition (EMT) is an important process for cancer metastasis, drug resistance, and cancer stem cells. Activation of fibroblast growth factor receptor 1 (FGFR1) was found to promote EMT and metastasis in prostate and breast cancers, but the effects and mechanisms in lung cancer was unclear. In this study, we aimed to explore whether and how activation of FGFR1 promotes EMT and metastasis in FGFR1-amplified lung cancer. We show that activation of FGFR1 by its ligand fibroblast growth factor 2 (FGF2) promoted proliferation, EMT, migration, and invasion in FGFR1-amplified lung cancer cell lines H1581 and DMS114, whereas inhibition of FGFR1 suppressed these processes. FGFR1 activation upregulated expression of Sry-related HMG box 2 (SOX2) by downstream phosphorylated ERK1/2; moreover, the upregulation of SOX2 by autophosphorylation variant ERK2_R67S plasmid transfection was not suppressed by FGFR1 inhibitor AZD4547 or MEK/ERK inhibitor AZD6244 in vitro. And SOX2 expression was also significantly upregulated in ERK2_R67S lentivirus-transfected stable cell lines in vivo. Overexpression of SOX2 promoted cell proliferation, EMT, migration, and invasion. Importantly, activation of FGFR1 could not promote these processes in SOX2-silenced stable cell lines. In orthotopic and subcutaneous lung cancer xenograft models, inhibition of FGFR1 suppressed tumor growth, SOX2 expression, EMT, and metastasis in vivo; however, these processes caused by SOX2-overexpressing stable cell lines were not suppressed by FGFR1 inhibition. Higher expression of FGFR1 and SOX2 were positively correlated, and both were associated with shorter survival in lung cancer patients. In conclusion, our findings reveal that activation of FGFR1 promotes cell proliferation, EMT, and metastasis by the newly defined FGFR1-ERK1/2-SOX2 axis in FGFR1-amplified lung cancer.
Journal Article
EGFR mutations induce the suppression of CD8+ T cell and anti-PD-1 resistance via ERK1/2-p90RSK-TGF-β axis in non-small cell lung cancer
Background
Non-small cell lung cancer (NSCLC) patients with
EGFR
mutations exhibit an unfavorable response to immune checkpoint inhibitor (ICI) monotherapy, and their tumor microenvironment (TME) is usually immunosuppressed. TGF-β plays an important role in immunosuppression; however, the effects of TGF-β on the TME and the efficacy of anti-PD-1 immunotherapy against
EGFR
-mutated tumors remain unclear.
Methods
Corresponding in vitro studies used the TCGA database, clinical specimens, and self-constructed mouse cell lines with
EGFR
mutations. We utilized C57BL/6N and humanized M-NSG mouse models bearing
EGFR
-mutated NSCLC to investigate the effects of TGF-β on the TME and the combined efficacy of TGF-β blockade and anti-PD-1 therapy. The changes in immune cells were monitored by flow cytometry. The correlation between TGF-β and immunotherapy outcomes of
EGFR
-mutated NSCLC was verified by clinical samples.
Results
We identified that TGF-β was upregulated in
EGFR
-mutated NSCLC by EGFR activation and subsequent ERK1/2-p90RSK phosphorylation. TGF-β directly inhibited CD8
+
T cell infiltration, proliferation, and cytotoxicity both in vitro and in vivo
,
but blocking TGF-β did not suppress the growth of
EGFR-
mutated tumors in vivo. Anti-TGF-β antibody combined with anti-PD-1 antibody significantly inhibited the proliferation of recombinant
EGFR
-mutated tumors in C57BL/6N mice, which was superior to their monotherapy. Mechanistically, the combination of anti-TGF-β and anti-PD-1 antibodies significantly increased the infiltration of CD8
+
T cells and enhanced the anti-tumor function of CD8
+
T cells. Moreover, we found that the expression of TGF-β1 in EGFR-TKI resistant cell lines was significantly higher than that in parental cell lines. The combination of anti-TGF-β and nivolumab significantly inhibited the proliferation of EGFR-TKI resistant tumors in humanized M-NSG mice and prolonged their survival.
Conclusions
Our results reveal that TGF-β expression is upregulated in NSCLC with
EGFR
mutations through the EGFR-ERK1/2-p90RSK signaling pathway. High TGF-β expression inhibits the infiltration and anti-tumor function of CD8
+
T cells, contributing to the “cold” TME of
EGFR
-mutated tumors. Blocking TGF-β can reshape the TME and enhance the therapeutic efficacy of anti-PD-1 in
EGFR
-mutated tumors, which provides a potential combination immunotherapy strategy for advanced NSCLC patients with
EGFR
mutations.
Graphical Abstract
Journal Article
Prediction of pathologic stage in non-small cell lung cancer using machine learning algorithm based on CT image feature analysis
Purpose
To explore imaging biomarkers that can be used for diagnosis and prediction of pathologic stage in non-small cell lung cancer (NSCLC) using multiple machine learning algorithms based on CT image feature analysis.
Methods
Patients with stage IA to IV NSCLC were included, and the whole dataset was divided into training and testing sets and an external validation set. To tackle imbalanced datasets in NSCLC, we generated a new dataset and achieved equilibrium of class distribution by using SMOTE algorithm. The datasets were randomly split up into a training/testing set. We calculated the importance value of CT image features by means of mean decrease gini impurity generated by random forest algorithm and selected optimal features according to feature importance (mean decrease gini impurity > 0.005). The performance of prediction model in training and testing sets were evaluated from the perspectives of classification accuracy, average precision (AP) score and precision-recall curve. The predictive accuracy of the model was externally validated using lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) samples from TCGA database.
Results
The prediction model that incorporated nine image features exhibited a high classification accuracy, precision and recall scores in the training and testing sets. In the external validation, the predictive accuracy of the model in LUAD outperformed that in LUSC.
Conclusions
The pathologic stage of patients with NSCLC can be accurately predicted based on CT image features, especially for LUAD. Our findings extend the application of machine learning algorithms in CT image feature prediction for pathologic staging and identify potential imaging biomarkers that can be used for diagnosis of pathologic stage in NSCLC patients.
Journal Article
Underwater Target Recognition Method Based on Singular Spectrum Analysis and Channel Attention Convolutional Neural Network
In order to improve the efficiency of the deep network model in processing the radiated noise signals of underwater acoustic targets, this paper introduces a Singular Spectrum Analysis and Channel Attention Convolutional Neural Network (SSA-CACNN) model. The front end of the model is designed as an SSA filter, and its input is the time-domain signal that has undergone simple preprocessing. The SSA method is utilized to separate the noise efficiently and reliably from useful signals. The first three orders of useful signals are then fed into the CACNN model, which has a convolutional layer set up at the beginning of the model to further remove noise from the signal. Then, the attention of the model to the feature signal channels is enhanced through the combination of multiple groups of convolutional operations and the channel attention mechanism, which facilitates the model’s ability to discern the essential characteristics of the underwater acoustic signals and improve the target recognition rate. Experimental Results: The signal reconstructed by the first three-order waveforms at the front end of the SSA-CACNN model proposed in this paper can retain most of the features of the target. In the experimental verification using the ShipsEar dataset, the model achieved a recognition accuracy of 98.64%. The model’s parameter count of 0.26 M was notably lower than that of other comparable deep models, indicating a more efficient use of resources. Additionally, the SSA-CACNN model had a certain degree of robustness to noise, with a correct recognition rate of 84.61% maintained when the signal-to-noise ratio (SNR) was −10 dB. Finally, the pre-trained SSA-CACNN model on the ShipsEar dataset was transferred to the DeepShip dataset with a recognition accuracy of 94.98%.
Journal Article
A Mean-Field Game Control for Large-Scale Swarm Formation Flight in Dense Environments
As an important part of cyberphysical systems (CPSs), multiple aerial drone systems are widely used in various scenarios, and research scenarios are becoming increasingly complex. However, planning strategies for the formation flying of aerial swarms in dense environments typically lack the capability of large-scale breakthrough because the amount of communication and computation required for swarm control grows exponentially with scale. To address this deficiency, we present a mean-field game (MFG) control-based method that ensures collision-free trajectory generation for the formation flight of a large-scale swarm. In this paper, two types of differentiable mean-field terms are proposed to quantify the overall similarity distance between large-scale 3-D formations and the potential energy value of dense 3-D obstacles, respectively. We then formulate these two terms into a mean-field game control framework, which minimizes energy cost, formation similarity error, and collision penalty under the dynamical constraints, so as to achieve spatiotemporal planning for the desired trajectory. The classical task of a distributed large-scale aerial swarm system is simulated by numerical examples, and the feasibility and effectiveness of our method are verified and analyzed. The comparison with baseline methods shows the advanced nature of our method.
Journal Article
A Rapid Prediction Method for Underwater Vehicle Radiated Noise Based on Feature Selection and Parallel Residual Neural Network
Efficient and high-precision prediction of underwater vehicle radiated noise is crucial for warship stealth assessment. To overcome the high modeling complexity and limited prediction capability of traditional methods, this paper proposes ADE-PNN-ResNet, a fast underwater radiated noise (URN) prediction model integrating Adaptive Differential Evolution (ADE) with a Parallel Residual Neural Network (PNN-ResNet). This data-driven framework replaces conventional physics-based modeling, significantly reducing complexity while preserving high prediction accuracy. This study includes three core points: Firstly, for each 1/3-octave target noise band, a joint feature selection strategy of measurement points and frequency bands based on the ADE is proposed to provide high-quality inputs for the subsequent model. Secondly, a Parallel Neural Network (PNN) is constructed by integrating Radial Basis Function Neural Network (RBFNN) that excels at handling local features and Multi-Layer Perceptron (MLP) that focuses on global features. PNN is then cascaded via residual connections to form PNN-ResNet, deepening the network layers and efficiently capturing the complex nonlinear relationships between vibration and noise. Thirdly, the proposed ADE-PNN-ResNet is validated using vibration and noise data collected from lake experiments of a scaled underwater vehicle model. Under the validation conditions, the absolute prediction error is below 3 dB for 96% of the 1/3-octave bands within the frequency range of 100–2000 Hz, with the inference time for prediction taking merely a few seconds. The research demonstrates that ADE-PNN-ResNet balances prediction accuracy and efficiency, providing a feasible intelligent solution for the rapid prediction of underwater vehicle radiated noise in engineering applications.
Journal Article