A5103110206- Nonlinear Differential Equations Analysis
- Nonlinear Partial Differential Equations
- Differential Equations and Numerical Methods
- Advanced Mathematical Physics Problems
- Machine Learning in Healthcare
- Advanced Mathematical Modeling in Engineering
- Nonlinear Waves and Solitons
- EEG and Brain-Computer Interfaces
- Topic Modeling
- Nonlinear Photonic Systems
- Bayesian Methods and Mixture Models
- Biomedical Text Mining and Ontologies
- Middle East Politics and Society
- Colorectal Cancer Screening and Detection
- Differential Equations and Boundary Problems
- Optimization and Variational Analysis
- Time Series Analysis and Forecasting
- Gaze Tracking and Assistive Technology
- Advanced Differential Equations and Dynamical Systems
- Emotion and Mood Recognition
- Numerical methods in engineering
- Pulmonary Hypertension Research and Treatments
- Artificial Intelligence in Healthcare and Education
- Fixed Point Theorems Analysis
- Fractional Differential Equations Solutions
Chinese Academy of Medical Sciences & Peking Union Medical College
2023-2025
Zhejiang University
2021-2023
Fu Wai Hospital
2023
Zhongkai University of Agriculture and Engineering
2012-2018
Yuxi Normal University
2004
Type 2 diabetes (T2D) is a worldwide chronic disease that difficult to cure and causes heavy social burden. Early prediction of T2D can effectively identify high-risk populations facilitate earlier implementation appropriate preventive interventions. Various early models for have been proposed. However, these methods do not consider the following factors: 1) health examination records (HER) containing information before diagnosis; 2) rating clinical knowledge; 3) local global time-series...
Survival analysis exhibits profound effects on health service management. Traditional approaches for survival have a pre-assumption the time-to-event probability distribution and seldom consider sequential visits of patients medical facilities. Although recent studies leverage merits deep learning techniques to capture non-linear features long-term dependencies within multiple analysis, lack interpretability prevents models from being applied clinical practice. To address this challenge,...
Massively available longitudinal data about long-term disease trajectories of patients provides a golden mine for the understanding progression and efficient health service delivery. It calls quantitative modeling progression, which is tricky problem due to complexity process as well irregularity time documented in trajectories. In this study, we tackle with goal predictively analyzing progression. Specifically, propose novel Variational Hawkes Process (VHP) model generalize predict future...
Personalized medicine requires the patient similarity analysis for providing specific treatments tailed each patient. However, in personalized clinical scenarios encounters challenges, which are twofold. First, heterogeneous and multi-type data usually recorded to Electronic Health Records (EHRs) during course of admissions, makes it difficult measure similarity. Second, disease progression manifests diverse states at different times, brings sequential complexity dynamically retrieve similar...
In this paper, we study the existence of multiple sign-changing solutions with a prescribed Lp+1−norm and theexistence least energy restrained for following nonlinear Schr¨odinger-Poisson system:−△u + u ϕ(x)u = λ|u|p−1u, in R3,−△ϕ(x) |u|2, R3.By choosing proper functional restricted on some appropriate subset to using method invariant sets descending flow,we prove that system has infinitely many forsuch solution p ∈ (3, 5). Few results...
Motivated by the non-stationarity characteristics of electroencephalograph (EEG) signals, we propose a signal variation elimination model (SVEM) for emotion recognition. The proposed SVEM enables to capture topological structures different EEG channels due utilized graph neural network (GNN). Two tricks are reduce variations and improve generalization. Firstly, is pre-trained mask-generation supervised learning where randomly mask several in GNN then generate them. Secondly, fine-tuned...