A5002822938- Advanced MRI Techniques and Applications
- Sparse and Compressive Sensing Techniques
- ECG Monitoring and Analysis
- EEG and Brain-Computer Interfaces
- Electrical and Bioimpedance Tomography
- Cardiac electrophysiology and arrhythmias
- Numerical methods in inverse problems
- AI in cancer detection
- Medical Image Segmentation Techniques
- Soil Moisture and Remote Sensing
- Image and Signal Denoising Methods
- Brain Tumor Detection and Classification
- Advanced Vision and Imaging
- Human Pose and Action Recognition
- Ultrasonics and Acoustic Wave Propagation
- Medical Imaging Techniques and Applications
- Photoacoustic and Ultrasonic Imaging
- Optical measurement and interference techniques
- Fault Detection and Control Systems
- Non-Invasive Vital Sign Monitoring
- Cancer Genomics and Diagnostics
- Structural Health Monitoring Techniques
- Blind Source Separation Techniques
- Image Processing Techniques and Applications
- Advanced Neural Network Applications
Zhejiang Sci-Tech University
2015-2025
City Hospital
2025
Harbin Medical University
2025
Third Affiliated Hospital of Harbin Medical University
2025
Guangdong Medical College
2025
Jiangsu University
2025
Ganzhou People's Hospital
2025
Shandong Agricultural University
2025
Hangzhou Cancer Hospital
2017-2024
Nanjing University of Finance and Economics
2023-2024
The commercial deployment of Aqueous Zinc Ion Batteries (AZIBs) is hampered by dendrites, the hydrogen evolution reaction (HER) and corrosion reactions. To tackle these challenges, we have introduced 3,3'‐dithiobis‐1‐propanesulfonic acid disodium salt (SPS), a symmetrical sulfur‐based organic salt, as an electrolyte additive for AZIBs. Unlike conventional additives that favour (002) deposition, SPS enables dense (100) growth through unique symmetrically aligned concentration‐controlled...
High-resolution magnetic resonance images can provide fine-grained anatomical information, but acquiring such data requires a long scanning time. In this paper, framework called the Fused Attentive Generative Adversarial Networks(FA-GAN) is proposed to generate super- resolution MR image from low-resolution images, which reduce time effectively with high images. of FA-GAN, local fusion feature block, consisting different three-pass networks by using convolution kernels, extract features at...
Research on undersampled magnetic resonance image (MRI) reconstruction can increase the speed of MRI imaging and reduce patient suffering. In this paper, an method based Generative Adversarial Networks with Self-Attention mechanism Relative Average discriminator (SARA-GAN) is proposed. our SARA-GAN, relative average theory applied to make full use prior knowledge, in which half input data true fake. At same time, a self-attention incorporated into high-layer generator build long-range...
<abstract> <p>Early diagnosis of abnormal electrocardiogram (ECG) signals can provide useful information for the prevention and detection arrhythmia diseases. Due to similarities in Normal beat (<italic>N</italic>) Supraventricular Premature Beat (<italic>S</italic>) categories imbalance ECG categories, classification cannot achieve satisfactory results under inter-patient assessment paradigm. In this paper, a multi-path parallel deep convolutional neural...
The reconstruction of epicardial potentials (EPs) from body surface (BSPs) can be characterized as an ill-posed inverse problem which generally requires a regularized numerical solution. Two kinds errors/noise: geometric errors and measurement exist in the ECG make solution such more difficulty. In particular, will directly affect calculation transfer matrix A linear system equation AX = B. this paper, we have applied truncated total least squares (TTLS) method to reconstruct EPs BSPs. This...
Automatic and accurate classification of Alzheimer's disease is a challenging promising task. Fully Convolutional Network (FCN) can classify images at the pixel level. Adding an attention mechanism to effectively improve performance model. However, self-attention ignores potential correlation between different samples. Aiming this problem, we propose new method for image based on external-attention mechanism. The module added after fourth convolutional block fully network At same time,...
Objective.With the improvement of living standards, heart disease has become one common diseases that threaten human health. Electrocardiography (ECG) is an effective way diagnosing cardiovascular diseases. With rapid growth ECG examinations and shortage cardiologists, accurate automatic arrhythmias classification a research hotspot. The main purpose this paper to improve accuracy in detecting abnormal patterns.Approach.A hybrid 1D Resnet-GRU method, consisting Resnet gated recurrent unit...
Liver function of chronic hepatitis B (CHB) patients is essentially normal after treatment with antiviral drugs. In rare cases, persistently abnormally elevated α-fetoprotein (AFP) seen in CHB following long-term treatment. However, the absence imaging evidence liver cancer, a reasonable explanation for this phenomenon still lacking. To explore causes abnormal AFP who were not diagnosed cancer. From November 2019 to May 2023, 15 and selected. Clinical data quality indicators related...
ABSTRACT Invasion and metastasis are major causes of mortality in breast cancer (BRCA) patients. LHPP, known for its tumor‐suppressive effects, has an undefined role BRCA. We found reduced LHPP protein BRCA tissues, with lower levels correlating poor patient outcomes. In vitro studies show inhibits cell proliferation, migration, invasion, stemness. vivo xenograft models support LHPP's curbing tumorigenesis lung metastasis. Mechanistically, interacts ERK P38 MAPK, leading to their...
Accurate drug–target binding affinity (DTA) prediction is crucial in drug discovery. Recently, deep learning methods for DTA have made significant progress. However, there are still two challenges: (1) recent models always ignore the correlations and target data drug/target representation process (2) interaction of pairs by simple concatenation, which insufficient to explore their fusion. To overcome these challenges, we propose an end-to-end sequence-based model called BTDHDTA. In feature...
Developing robust Alzheimer's Disease (AD) classification models necessitates extensive training data, but aggregating multi-center medical data poses privacy risks. Although Federated Learning (FL) and Swarm (SL) allow generic without sharing, their performance is limited by variations in AD pathology features sample class imbalances across centers. To address this issue, we propose a novel Harmonized framework with Guided Optimization (HSGO) to enhance collaboration while preserving...