A5065150622- Advanced MRI Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
- Advanced Neuroimaging Techniques and Applications
- Lanthanide and Transition Metal Complexes
- Electron Spin Resonance Studies
- MRI in cancer diagnosis
- Radiomics and Machine Learning in Medical Imaging
- Spectroscopy Techniques in Biomedical and Chemical Research
- Random lasers and scattering media
- Image and Signal Denoising Methods
- Natural Language Processing Techniques
- Advanced Adaptive Filtering Techniques
- Nuclear Physics and Applications
- Control Systems and Identification
- Seismic Imaging and Inversion Techniques
- Color perception and design
- Face and Expression Recognition
- Blind Source Separation Techniques
- Medical Imaging Techniques and Applications
- Emotion and Mood Recognition
Guizhou University
2024
Mitsubishi Electric (Japan)
2023
Xiamen University
2022-2023
State Grid Corporation of China (China)
2021
Jilin Electric Power Research Institute (China)
2021
Anhui Polytechnic University
2021
North China Electric Power University
2021
Hefei University
2021
Hainan University
2021
To develop a deep learning-based method, dubbed Denoising CEST Network (DECENT), to fully exploit the spatiotemporal correlation prior image denoising.DECENT is composed of two parallel pathways with different convolution kernel sizes aiming extract global and spectral features embedded in images. Each pathway consists modified U-Net residual Encoder-Decoder network 3D convolution. Fusion 1 × utilized concatenate pathways, output DECENT noise-reduced The performance was validated numerical...
Abstract Chemical exchange saturation transfer (CEST) is a versatile technique that enables noninvasive detections of endogenous metabolites present in low concentrations living tissue. However, CEST imaging suffers from an inherently signal‐to‐noise ratio (SNR) due to the decreased water signal caused by saturated spins. This limitation challenges accuracy and reliability quantification imaging. In this study, novel spatial–spectral denoising method, called BOOST (suBspace with nOnlocal...
Emotional recognition is a pivotal research domain in computer and cognitive science. Recent advancements have led to various emotion methods, leveraging data from diverse sources like speech, facial expressions, electroencephalogram (EEG), electrocardiogram, eye tracking (ET). This article introduces novel framework, primarily targeting the analysis of users’ psychological reactions stimuli. It important note that stimuli eliciting emotional responses are as critical themselves. Hence, our...
Purpose This work introduces and validates a deep‐learning‐based fitting method, which can rapidly provide accurate robust estimation of cytological features brain tumor based on the IMPULSED (imaging microstructural parameters using limited spectrally edited diffusion) model with diffusion‐weighted MRI data. Methods The U‐Net was applied to quantify extracellular diffusion coefficient ( D ex ), cell size d intracellular volume fraction v in ) tumor. At training stage, image‐based data,...
Purpose To present a deep learning–based reconstruction method for spatiotemporally encoded single‐shot MRI to simultaneously obtain water and fat images. Methods Spatiotemporally is an ultrafast branch that can encode chemical shift information due its special quadratic phase modulation. A learning approach using 2D U‐Net was proposed reconstruct signal images simultaneously. The training data were generated by MRiLab software (version 1.3) with various synthetic models. Numerical...
Most of the water/fat separation techniques need to acquire multiple images with different echo time, which usually take long acquisition time. Multiple overlapping-echo detachment (MOLED) imaging can shorten time by acquiring MR signals in same k-space. Here, a new method for fast T2* mapping MOLED technology was proposed, obtain quantitative maps and M0 water fat single shot. In vivo experiment demonstrates that proposed accurate parameter values images.
Chemical exchange saturation transfer (CEST) is a powerful technique that enables non-invasive detection of endogenous metabolites in living tissues. Since the observed water signal decreased due to saturated spins, CEST imaging inherently suffers from low SNR, hence degrading accuracy and reproducibility. Inspired by spatial-spectral correlation images, here we propose Subapace denoising method with Non-Local Low-Rank constraint Spectral-Smoothness regularization (SNLRSS) diminish noise,...
Motivation: Skeletal muscle inflammation/necrosis and fat infiltration are strong indicators of disease activity progression in many neuromuscular disorders. They can be assessed by T2 relaxometry water-fat separation techniques, respectively. Goal(s): Develop a method for simultaneous quantification. Approach: The chemical-shift encoding multiple overlapping-echo detachment (CSE-MOLED) sequence was designed MRI data acquisition, synthetic deep learning were used image reconstruction....
This study assesses the feasibility of training a convolutional neural network (CNN) for IMPULSED (imaging microstructural parameters using limited spectrally edited diffusion) model fitting to diffusion-weighted (DW) data and evaluates its performance on brain tumor (poorly differentiated adenocarcinoma) patient directly acquired from clinical MR scanner. Comparisons were made with results calculated non-linear least squares (NLLS) algorithm. More accurate robust obtained by our CNN method,...
The DXHS algorithm was proposed as an active noise control method and is often referred to adaptive method. This paper shows that the a dynamical system equivalent passive internal model for sinusoidal signals. It follows disturbance attenuation ability consequence of principle, although based on standard way thinking methods. Moreover, closed-loop stability guaranteed strictly secondary paths.
Water-fat separation is a powerful tool in diagnosing many diseases and efforts have been made to reduce the scan time. Spatiotemporally encoded (SPEN) single-shot MRI, as an emerging ultrafast MRI method, can accomplish fastest water-fat since only one shot required. However, SPEN water/fat images obtained by state-of-the art methods still some shortcomings. Here, deep learning approach based on U-Net was proposed obtain simultaneously with improved spatial resolution, better fidelity...