A5048902633- Speech Recognition and Synthesis
- Speech and Audio Processing
- ECG Monitoring and Analysis
- Phonetics and Phonology Research
- Cardiac electrophysiology and arrhythmias
- Cardiac Imaging and Diagnostics
- EEG and Brain-Computer Interfaces
- Heart Rate Variability and Autonomic Control
- Non-Invasive Vital Sign Monitoring
- Fluid Dynamics and Turbulent Flows
- Iterative Learning Control Systems
- Speech and dialogue systems
- RNA modifications and cancer
- Music and Audio Processing
- Fault Detection and Control Systems
- Machine Fault Diagnosis Techniques
- Turbomachinery Performance and Optimization
- Complex Systems and Time Series Analysis
- Adaptive Dynamic Programming Control
- Cardiac tumors and thrombi
- Multisensory perception and integration
- Computational Fluid Dynamics and Aerodynamics
- Gait Recognition and Analysis
- Artificial Intelligence in Healthcare
- Face recognition and analysis
China International Science and Technology Cooperation
2025
Zhejiang Chinese Medical University
2025
Shandong University
2022-2024
China Telecom (China)
2024
China Telecom
2024
Yunnan University
2024
South China University of Technology
2016-2021
Guangdong Academy of Medical Sciences
2021
Guangdong Provincial People's Hospital
2021
Hitachi (Japan)
2008-2017
Purpose: Acute myocardial infarction (AMI) is a major contributor to death. The purpose of this study explore circulating biomarkers for AMI diagnosis from the perspectives immunological microenvironment and N6-methyladenosine (m6A) RNA methylation regulation. Patients Methods: GSE59867 dataset was used download platform probe data conducting differential analysis m6A regulators. A diagnostic nomogram created utilizing random-forest method evaluated predictive power. m6A-related gene...
Glucagon-like peptide-1 receptor (GLP-1R), glucose-dependent insulinotropic polypeptide (GIPR), and G protein-coupled 40 (GPR40) are members of receptors (GPCR) family. They abundantly expressed in islet β cells, mediate effects incretins fatty acids cells. Glucose 5-AMP-activated protein kinase (AMPK) known to be involved the regulation cell function. Metformin potential therapeutic drug for type 2 diabetes, 5-amino-4-imidazolecarboxamide riboside (AICAR), both activators AMPK. Here we...
Cardiovascular disease disturbs the structure and function of myocardial cells, altering spatiotemporal dynamic patterns cardiac electrical activity. The conventional joint recurrence analysis contributes to revealing pattern heart disease-induced changes, but it usually overlooks intrinsic dynamics heart. In clinical practice, multiple non-cardiovascular factors can cause similar alterations in heart's activity that those observed cardiovascular disease. Therefore, accurately diagnosing...
Abstract Objective. Myocardial infarction (MI) is one of the leading causes human mortality in all cardiovascular diseases globally. Currently, 12-lead electrocardiogram (ECG) widely used as a first-line diagnostic tool for MI. However, visual inspection pathological ECG variations induced by MI remains great challenge cardiologists, since changes are usually complex and slight. Approach. To have an accuracy detection, prominent features extracted from in-depth mining signals need to be...
Abstract Background The cardiodynamicsgram (CDG), a novel noninvasive method, extracts dynamic ST‐T segment information from an electrocardiogram (ECG) through deterministic learning. Hypothesis CDG can reflect anomalous functional in coronary artery disease (CAD). Methods We retrospectively enrolled 456 patients with suspected CAD who underwent computed tomography angiography (CCTA) January 2020 to 2022, followed immediately by standard 12‐lead ECG acquisition. Positivity for were defined...
Background: Electrocardiogram (ECG) is a routine method for detecting myocardial ischemia in clinical practice, but more than half of ECGs are without specific ischemic changes. Cardiodynamicsgram (CDG) an effective to detect with non-diagnostic ECG. The Lyapunov exponent (LYE) and the Fourier transform coefficient combined characterize spatial temporal features CDG. However, some cases, does not accurately enough describe degree irregular morphology CDG patients. In this context, study aims...
A new method for generating sentence F 0 contours of Mandarin speech is proposed.The assumes the contour generation process model, but generates tone and phrase components in different ways sums them to produce a contour.The component generated concatenating patterns nuclei, which are predicted by corpus-based scheme (binary decision trees).Experiments were conducted using 100 news utterances female speaker.The results showed that could generate F0 close those target speech.A perceptual...
In this paper, a rule-based method was developed for realizing phrase components in our two-step generation of fundamental frequency (F0) contours Mandarin.The scheme assumes (logarithmic) F0 as superposition tone on components, which are further assumed to be responses commands.In general, possibility new command comes higher at deeper syntactic boundaries, but is also affected by the distance from preceding command.A long interval causes flat contour close baseline, not case human...
Abstract Objective. Myocardial infarction (MI) is a prevalent cardiovascular disease that contributes to global mortality rates. Timely diagnosis and treatment of MI are crucial in reducing its fatality rate. Currently, electrocardiography (ECG) serves as the primary tool for clinical diagnosis. However, detecting accurately through ECG remains challenging due complex subtle pathological changes it causes. To enhance accuracy MI, more thorough exploration signals necessary extract...
Abstract Objective. In recent years, artificial intelligence-based electrocardiogram (ECG) methods have been massively applied to myocardial infarction (MI). However, the joint analysis of static and dynamic features achieve accurate interpretable MI detection has not comprehensively addressed. Approach. This paper proposes a simplified ensemble tree method with solve this issue for detection. Initially, are extracted by modeling intrinsic dynamics ECG via learning in addition extracting...
Flexible sensors as emerging wearable electronics are widely utilized in exercise guidance, medical services, intelligent robots, etc., providing rich data for the Internet of Things and promoting intelligence convenience people's lives. However, it is still difficult to design fabricate multifunctional flexible with excellent comprehensive performance universality by balancing various sensing parameters a scientific economical way. Herein, sensor synergistic conductive network composed...
Improved prediction of tone components was realized in our method for synthesizing sentence fundamental frequency (F 0 ) contours Mandarin speech.The is based on representing a logarithmic F contour as superposition phrase the case generation process model model).The are by concatenating their fragments at nuclei predicted corpus-based method, while generated rules under framework.In original assumed to have similar shapes nuclei.This assumption that almost flat throughout an...
In this paper we propose a new cross-lingual Voice Conversion (VC) approach which can generate all speech parameters (MCEP, LF0, BAP) from one DNN model using PPGs (Phonetic PosteriorGrams) extracted inputted several ASR acoustic models. Using the proposed VC method, tried three different approaches to build multilingual TTS system without recording corpus. A listening test was carried out evaluate both quality (naturalness) and voice similarity between converted target speech. The results...