A5100343540- Electrical and Bioimpedance Tomography
- Non-Destructive Testing Techniques
- Biometric Identification and Security
- Body Composition Measurement Techniques
- Face and Expression Recognition
- Geophysical and Geoelectrical Methods
- Image and Video Stabilization
- EEG and Brain-Computer Interfaces
- Medical Image Segmentation Techniques
- Neural Networks and Applications
- AI in cancer detection
- Hemodynamic Monitoring and Therapy
- Flow Measurement and Analysis
- Magnetic Field Sensors Techniques
- Non-Invasive Vital Sign Monitoring
- Blind Source Separation Techniques
- Microfluidic and Bio-sensing Technologies
- Muscle activation and electromyography studies
- Radiomics and Machine Learning in Medical Imaging
- Characterization and Applications of Magnetic Nanoparticles
- Image and Signal Denoising Methods
- Face recognition and analysis
- Model Reduction and Neural Networks
- Emotion and Mood Recognition
- Asthma and respiratory diseases
Shenyang University of Technology
2015-2025
Southeast University
2003-2025
Zhengzhou University of Light Industry
2025
Xiamen University
2024
New York Medical College
2023
Guangzhou University of Chinese Medicine
2023
Nanchang University
2016-2022
First Affiliated Hospital of Nanchang University
2022
Northeastern University
2008-2010
Maine Medical Center
2004
aims: To improve the accuracy of wind power prediction background: In recent years, has developed rapidly nationwide and worldwide.How to precision become a research hotspot in this field objective: This paper aims construct forecasting system based on SENet-CNN-LSTM stability method: Firstly, isolated forest algorithm is used detect abnormal values historical data, linear interpolation method fill missing data. Secondly, CNN extract spatio-temporal characteristics SENet assign different...
This paper proposes a Kolmogorov high order deep neural network (K-HOrderDNN) for solving high-dimensional partial differential equations (PDEs), which improves the networks (HOrderDNNs). HOrderDNNs have been demonstrated to outperform conventional DNNs frequency problems by introducing nonlinear transformation layer consisting of $(p+1)^d$ basis functions. However, number functions grows exponentially with dimension $d$, results in curse dimensionality (CoD). Inspired superposition theorem...
This report introduces Dolphin, a large-scale multilingual automatic speech recognition (ASR) model that extends the Whisper architecture to support wider range of languages. Our approach integrates in-house proprietary and open-source datasets refine optimize Dolphin's performance. The is specifically designed achieve notable accuracy for 40 Eastern languages across East Asia, South Southeast Middle East, while also supporting 22 Chinese dialects. Experimental evaluations show Dolphin...
Rationale IgE plays an important pathologic role in most, if not all, allergic conditions. We previously showed that ASHMI (anti-asthma herbal medicine intervention) suppressed production murine models of asthma and subjects. However, the active compounds responsible for suppression are still unknown. Objective sought to identify compound(s) inhibition as well investigate mechanisms by which identified decreases production. Methods The Sophorae Flavescentis were separated using Column...
Computer-aided diagnosis (CAD) system can promote the detection accuracy by providing a "second opinion" to radiologist, so high of mass in mammogram is critical for improving performance and efficiency. In this paper, we designed auto-diagnosis method based on texture features. First, was detected base bilateral comparison, center region interest (ROI) located. Second, fractal dimension two-dimensional entropy were calculated as Last, kinds ROI diagnosed Support Vector Machine (SVM), or...
Background: Cerebral blood flow (CBF) auto-regulation function is essential for evaluating CBF status. This article provides an evaluation method of using bioimpedance. Method: A mathematical model cerebral vascular compliance established to analyze the effect autoregulation ability changes on impedance CBF. The synchronization signals and pressure are measured, peak phase shift extracted evaluate function. Results: experiment results show that can express change process. frontal occipital...
Signal processing of Electroencephalography (EEG) plays an important role in brain-computer-interface (BCI) system. Therefore, it is a key selection suitable methods. This paper proposed new method by combing Independent Component Analysis (ICA) with One Versus the Rest Common Spatial Patterns (OVR-CSP) to improve classification performance. Firstly, EEG signals were filtered FIR bandpass filter 8-30HZ. Secondly, relative frequency band (i.e. μ and β rhythm) decomposed into independent...
Magnetic induction tomography (MIT) acted as a contactless and non-invasive medical imaging technology has aroused wide concern, while large amount of calculation series convergence problems in the solution inverse problem become technical difficulties for MIT. In order to solve these problems, an improved back-projection image reconstruction algorithm based on magnetic field lines distribution is presented this paper. Firstly, eddy current MIT was solved by finite element method obtain...
Facial expression recognition (FER) plays a significant part in artificial intelligence and computer vision. However, most of facial methods have not obtained satisfactory results based on low‐level features. The existed used encountered the major issues linear inseparability, large computational burden, data redundancy. To obtain results, we propose an innovative deep learning (DL) model using kernel entropy component analysis network (KECANet) directed acyclic graph support vector machine...
In order to denoise image and improve its visual quality, an improved Wiener filtering method is proposed based on wavelet transform. First, noise analyzed, then the corrupted by given. The noisy denoised procession repeated until satisfied requirements. Experiment results demonstrate that new removes of improves with PSNR between original increased 2.02, while similar image. It can get conclusion good effect, it effective processing method.
Tau pathology, including aberrant tau hyperphosphorylation, aggregation, and mislocalization, is implicated in many neurodegenerative disorders, Alzheimer’s disease (AD), which the most prevalent dementia among elderly. A better understanding of molecular mechanisms underlying pathology should help advance therapies for diseases. Perturbations cyclic adenosine monophosphate (cAMP)-dependent signaling play an important role pathophysiology numerous neurological EPAC2 intracellular...
Eyes location is an important basis for a face recognition system. In this paper, we proposed novel method to improve the eyes accuracy in image, which detected by combining dual-orientation Gabor filters and eye templates (left right) , were got from average of 20 images collected randomly database, eyes. First, YCrCb color space was used extract region then get interest regions human filters. Finally, matched with position located accurately. The experimental results on FERET database...
Magnetic induction tomography (MIT) is a tomographic imaging technique, which has potential applications in security, industry, and medicine. Typically, sensors form closed structure around the object. However, measurement cannot be achieved using sensor array process of severe brain trauma nursing neurosurgery operation.The new sector magnetic (SMIT) system developed to realize real-time monitoring treatment brain. The functions drive coil are separated this system. detection sensitivity...
Magnetic induction tomography (MIT) is a new biologic technology with the feature of harmless, non-invasive and convenience. The resolution speed image reconstruction algorithm critical for improving performance MIT system its application. Newton-One-Step Error Reconstruct or (NOSER) common in MIT, but slight variations original data will impact reconstructed images because Hessian matrix which ill-posed process NOSER. In this paper, NOSER was improved by eigen value threshold(ET) used to...