A5100680886- Sparse and Compressive Sensing Techniques
- Stochastic Gradient Optimization Techniques
- Evacuation and Crowd Dynamics
- Complexity and Algorithms in Graphs
- Childhood Cancer Survivors' Quality of Life
- Family Support in Illness
- Privacy-Preserving Technologies in Data
- Epigenetics and DNA Methylation
- Face and Expression Recognition
- Cryptography and Data Security
- Markov Chains and Monte Carlo Methods
- Seismic Imaging and Inversion Techniques
- Advanced Bandit Algorithms Research
- Ubiquitin and proteasome pathways
- Advanced Differential Equations and Dynamical Systems
- Menstrual Health and Disorders
- Machine Learning and Algorithms
- Game Theory and Voting Systems
- Artificial Intelligence in Healthcare and Education
- Mitochondrial Function and Pathology
- Chaos control and synchronization
- Coastal wetland ecosystem dynamics
- Tensor decomposition and applications
- Enhanced Oil Recovery Techniques
- Cellular and Composite Structures
Shanghai Jiao Tong University
2014-2025
Huazhong University of Science and Technology
2019-2024
St. Jude Children's Research Hospital
2022-2024
Karlsruhe Institute of Technology
2024
State Key Laboratory of Oncogene and Related Genes
2024
Nanjing University of Aeronautics and Astronautics
2023
Shanghai Institute of Technology
2022
Nanjing Hydraulic Research Institute
2021
Xi'an University of Science and Technology
2021
Hohai University
2021
Junctophilin-2 (JP2) is a structural protein required for normal excitation-contraction (E-C) coupling. After cardiac stress, JP2 cleaved by the calcium ion-dependent protease calpain, which disrupts E-C coupling ultrastructural machinery and drives heart failure progression. We found that stress-induced proteolysis of liberates an N-terminal fragment (JP2NT) translocates to nucleus, binds genomic DNA, controls expression spectrum genes in cardiomyocytes. Transgenic overexpression JP2NT mice...
Despite the high effectiveness of HPV vaccines in preventing infection, vaccine hesitancy remains a concern, particularly China. This study aimed to explore college students' attitudes toward vaccination and identify associated factors. Data was collected through cross-sectional survey using self-administered questionnaires four cities from May June 2022. Chi-square tests logistic regression analyses were conducted Additionally, an integrated structural equation model (SEM) based on 3Cs...
To investigate the prevalence of vitamin D deficiency (VDD) in children/adolescents extreme southern China. This multicenter, cross-sectional study included 21,811 children aged 0-18 years from 18 districts Hainan Province, using a multistage stratified random sampling method January 2021 to March 2022. Serum 25(OH)D levels decreased with age (p trend <0.001). VDD increased significantly 3.7% (95% CI: 3.2, 4.3) 0-3 43.5% 42.1, 45.0) those 13-18 years. Girls and urban residents showed higher...
This study aimed to investigate the impact of tibial anatomical structure, age, sex, average weekly exercise time during adolescence (AWETA), and BMI on risk stress fractures (SF) develop a predictive model using logistic regression. We retrospectively analyzed 748 patients presenting with calf pain at our hospital from January 1, 2018, August 31, 2023. After applying inclusion exclusion criteria, 493 were categorized into SF group (295 cases) control (198 cases). Detailed patient...
Abstract Drought is a critical environmental challenge affecting plant growth and productivity. Understanding the regulatory networks governing drought response at cellular level remains an open question. Here, comprehensive multi‐omics integration framework that combines transcriptomic, proteomic, epigenetic, network‐based analyses to delineate cell‐type‐specific involved in presented. By analyzing nearly 30 000 data samples across species, unique insights are revealed into conserved...
Abstract Deep learning has been used extensively in histopathological image classification, but people this field are still exploring new neural network architectures for more effective and efficient cancer diagnosis. Here, we propose multi-scale, multi-view progressive feature encoding (MSMV-PFENet) classification. With respect to the density of cell nuclei, selected regions potentially related carcinogenesis at multiple scales from each view. The then extracted global local features these...
Abstract Background Childhood cancer survivors are at high risk for morbidity and mortality poor patient-reported outcomes, typically health-related quality of life (HRQOL). However, associations between DNA methylation–based aging biomarkers HRQOL have not been evaluated. Methods methylation was generated with Infinium EPIC BeadChip on blood-derived (median age blood draw = 34.5 years, range 18.5-66.6 years), assessed survey (mean 32.3 18.4-64.5 years) from 2206 in the St Jude Lifetime...
In this paper, we characterize the dynamics of Chen system ẋ = a(y - x), ẏ (c a)x xz + cy, ż xy bz which has an invariant algebraic surface.
We consider saddle point problems which objective functions are the average of $n$ strongly convex-concave individual components. Recently, researchers exploit variance reduction methods to solve such and achieve linear-convergence guarantees. However, these have a slow convergence when condition number problem is very large. In this paper, we propose stochastic proximal algorithm, accelerates method SAGA for problems. Compared with catalyst framework, our algorithm reduces logarithmic term...
K-anonymity is a popular model used in microdata publishing to protect individual privacy. This paper introduces the idea of ball tree and projection area density partition into k-anonymity algorithm.The traditional kd-tree implements division by forming super-rectangular, but super-rectangular has angle, so it cannot guarantee that records on corner are most similar this area. In paper, super-sphere formed ball-tree address problem. We adopt increase resulting recorded points. implement our...
Abstract This paper investigates the application of unsupervised learning methods for computed tomography reconstruction. To motivate our work, we review several existing priors, namely truncated Gaussian prior, total variation and deep image prior (DIP). We find that DIP outperforms other three priors in terms representational capability visual performance. However, performance deteriorates when number iterations exceeds a certain threshold due to overfitting. address this issue, propose...
Abstract Shock loading and damage to drill bit cutters caused by changes in rock formations has long been a challenge, especially for shoulder that are subjected the highest cutting forces. To overcome this problem, breakthrough solution is emerging—the "Adaptive" polycrystalline diamond compact (PDC) bit, designed with mounted onto specially engineered elastic structure. This innovation aims mitigate damaging vibrations, extend lifespan, improve penetration rates highly heterogeneous...
Summary Micromodels play a significant role in investigating flow, transport, interaction of various substances within synthetically fabricated micro-spaces, typically constructed from artificial materials such as PDMS, glass and silicon. also show substantial potential for addressing critical challenges the energy transition, e.g., CO2 or hydrogen flow properties, mineral dissolution/precipitation, microbial activities porous rocks. However, rarity real-rock micromodels that incorporate...
The linear contextual bandits is a sequential decision-making problem where an agent decides among actions given their corresponding contexts. Since large-scale data sets become more and common, we study the in high-dimensional situations. Recent works focus on employing matrix sketching methods to accelerating bandits. However, approximation error will bring additional terms regret bound. In this paper first propose novel method which called Spectral Compensation Frequent Directions (SCFD)....
Predicting the behaviors of pedestrian crowds is critical importance for a variety real-world problems. Data driven modeling, which aims to learn mathematical models from observed data, promising tool construct that can make accurate predictions such systems. In this work, we present data-driven modeling approach based on ODE-Net framework, constructing continuous-time crowd dynamics. We discuss some challenging issues in applying method problems, are primarily associated with dimensionality...
We study the streaming model for approximate matrix multiplication (AMM). are interested in scenario that algorithm can only take one pass over data with limited memory. The state-of-the-art deterministic sketching AMM is co-occurring directions (COD), which has much smaller approximation errors than randomized algorithms and outperforms other methods empirically. In this paper, we provide a tighter error bound COD whose leading term considers potential low-rank structure correlation of...