A5071153790- Non-Invasive Vital Sign Monitoring
- Heart Rate Variability and Autonomic Control
- ECG Monitoring and Analysis
- Advanced Computational Techniques and Applications
- Advanced Algorithms and Applications
- Advanced Measurement and Detection Methods
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Remote-Sensing Image Classification
- Optical Systems and Laser Technology
- Industrial Technology and Control Systems
- Engineering and Test Systems
- Industrial Automation and Control Systems
- Handwritten Text Recognition Techniques
- Remote Sensing and Land Use
- Image Processing and 3D Reconstruction
- Smart Grid and Power Systems
- EEG and Brain-Computer Interfaces
- Emotion and Mood Recognition
- Advanced Computing and Algorithms
- Advanced SAR Imaging Techniques
- Advanced Image Processing Techniques
- Advanced Vision and Imaging
- Cardiovascular Health and Disease Prevention
- Advanced Sensor and Control Systems
- Infrared Target Detection Methodologies
Suzhou Institute of Biomedical Engineering and Technology
2021-2024
Chinese Academy of Sciences
2021-2024
Xidian University
2018-2021
Southwest China Institute of Electronic Technology
2010-2011
Southwestern Institute of Physics
2011
Beihang University
2005
Non-invasive heart rate estimation is of great importance in daily monitoring cardiovascular diseases. In this paper, a bidirectional long short term memory (bi-LSTM) regression network developed for non-invasive from the ballistocardiograms (BCG) signals. The proposed deep model provides an effective solution to existing challenges BCG estimation, such as mismatch between signals and ground-truth reference, multi-sensor fusion time series feature learning. Allowing label uncertainty can...
Inspired by the application of recurrent neural networks (RNNs) to image recognition, in this paper, we propose a heartbeat detection framework based on Gated Recurrent Unit (GRU) network. In contribution, task from ballistocardiogram (BCG) signals was modeled as classification problem where segments BCG were formulated images fed into GRU network for feature extraction. The proposed has advantages fusion multi-channel and effective extraction temporal waveform characteristics signal,...
In recent years, research on human psychological stress using wearable devices has gradually attracted attention. However, the physical and differences among individuals high cost of data collection are main challenges for further this problem. work, our aim is to build a model detect subjects’ in different states through electrocardiogram (ECG) signals. Therefore, we design VR high-altitude experiment induce subject obtain ECG signal dataset. experiment, participants wear smart T-shirts...
A novel classification method for polarimetric synthetic aperture radar (PolSAR) images using support vector machine (SVM) with self-paced learning (SPL) optimization is proposed in the paper. In our method, Cloude-Pottier decomposition components and eigenvalues of coherency matrix are used as features. Classification carried out SVM, SPL to improve classifier achieve a stronger generalization capacity. Under paradigm, learns easier samples first gradually involves more difficult into...
A simple and fast approach based on eigenvalue similarity metric for Polarimetric SAR image segmentation of Land Cover is proposed in this paper. The uses eigenvalues the coherency matrix as to construct clustering algorithm segment image. Mahalanobis distance used pairwise between pixels avoid manual scale parameter tuning previous spectral method. Furthermore, spatial coherence constraints ensemble are employed stabilize improve performance. All experiments carried out three sets data....
Heart rate is one of the most important diagnostic bases for cardiovascular disease. This paper introduces a deep autoencoding strategy into feature extraction electrocardiogram (ECG) signals, and proposes beat-to-beat heart estimation method based on convolution Gaussian mixture clustering. The high-level heartbeat features were first extracted in an unsupervised manner by training convolutional autoencoder network, then adaptive clustering was applied to detect locations from features,...
In order to address the issue of heart rate susceptibility motion artifacts (MAs) when extracting it from photoplethysmography (PPG) signals, a estimation algorithm based on finite state machine (FSM) is proposed. The first applies band-pass filtering PPG and three-axis acceleration signals. strength MA assessed data. If strong detected, recursive least squares (RLS) applied; otherwise, omitted. Then, signal subjected an empirical wavelet transform (EWT). Based EWT results, current...