A5103189533- Radiomics and Machine Learning in Medical Imaging
- MRI in cancer diagnosis
- Thyroid Cancer Diagnosis and Treatment
- Medical Image Segmentation Techniques
- Advanced MRI Techniques and Applications
- Advanced Neural Network Applications
- Advanced X-ray and CT Imaging
- AI in cancer detection
- Voice and Speech Disorders
- Lung Cancer Diagnosis and Treatment
- Microbial Applications in Construction Materials
- Cerebrovascular and Carotid Artery Diseases
- Grouting, Rheology, and Soil Mechanics
- IgG4-Related and Inflammatory Diseases
- Stochastic Gradient Optimization Techniques
- Medical Imaging Techniques and Applications
- Advanced Radiotherapy Techniques
- Acute Ischemic Stroke Management
- Bluetooth and Wireless Communication Technologies
- Cardiovascular Health and Disease Prevention
- Endometriosis Research and Treatment
- Head and Neck Cancer Studies
- Explainable Artificial Intelligence (XAI)
- Reproductive System and Pregnancy
- Millimeter-Wave Propagation and Modeling
Shandong University
2025
Guizhou Minzu University
2024
Nanjing University of Chinese Medicine
2024
La Consolacion University Philippines
2024
Shanghai University
2015-2021
University of Science and Technology Beijing
2021
Qingdao University
2011
Qingdao Municipal Hospital
2011
The time-varying viscosity of quick-setting slurries will cause the spatial variation slurry viscosity, which has an important impact on diffusion mechanism infiltration grouting. With Bingham fluid constitutive model and uniform capillary group theory, a cylindrical–hemispherical was established, considering time-space viscosity. According to soil parameters rheological parameters, pressure can be calculated at different radius times. effects permeability, porosity, grouting rate injected...
This study intends to determine the expectations of teachers as well experiences that they have had while teaching computer music composition in a selection colleges and universities located province Hubei China. In order gather perception particular respondent detect structured pattern, this will be based on descriptive correlational research design. A numerical score rating issued each respondent. According Health Research Funding (2018), quantitative enables researchers objective, collect...
The ultrasonic diagnosis of lymph node lesions is usually based on a small number subjective visual features from single modality, which limits diagnostic accuracy. Therefore, our study aimed to propose computerized method for using dual-mode ultrasound radiomics and the intrinsic imaging phenotypes accurately differentiating benign, lymphomatous, metastatic nodes.A total 543 nodes 538 patients were examined with both B-mode ultrasonography elastography. data set was randomly divided into...
Due to it is difficulty segment the steel bar image in a complex background and with external interference. In this paper, we propose multi-attention U-Net image. First of all, order accurately find region filter noise part, add row mean attention module decoding path U-shaped network by using characteristic constant diameter bar, which reduces be segmented into bar. addition, an branch optimization strategy based on mask designed, uses output high-level semantics features adjacent low-level...
This study aimed to evaluate MAGiC imaging (one synthetic MRI technique, syMRI) and its combination with diffusion-weighted (DWI) in discriminating benign from metastatic retropharyngeal lymph nodes (RLNs). With derived relaxation parameters, 58 patients 21 42 RLNs were measured. The resultant T1, T2, PD T1SD values showed significant different between an optimal diagnostic performance T1SD. Moreover, the of MAGiC, DWI, morphological features demonstrated a significantly improved on overall...
It's very meaningful to conduct the driver parking space available in lot clearly and accurately by computer vision computational intelligence. While it is an extremely difficult task, because of scene complexity lot. In this paper, a new intelligence technology proposed detect spaces outdoor based on fusing Region Convolution Neural Network (RCNN) random forests (RF), we called as RCNN-RF. Within RCNN-RF, activation function design. And with help forests, RCNN-RF overcomes includes...
Federated Learning (FL) has emerged as a popular paradigm for collaborative learning among multiple parties. It is considered privacy-friendly because local data remains on personal devices, and only intermediate parameters -- such gradients or model updates are shared. Although gradient inversion widely viewed common attack method in FL, analytical research reconstructing input training samples from shared limited typically confined to constrained settings like small batch sizes. In this...
Motivation: Synthetic-DWI (SyDWI) at high-b-value, derived from low-b-value DWI, might be beneficial for prostate cancer evaluation due to better conspicuity of lesions. Relative conventional reconstruction (ConR), a vendor-provided deep learning (DLR) has been reported improving imaging quality in aspects higher SNR and sharpness. Goal(s): Investigate the impact DLR on image SyDWI lesion detection. Approach: Low-b-value DWI was reconstructed with ConR, separately. SyDWIs were generated...
A common approach when reconstructing a 3D model from medical images is to do segmentation and registration manually reconstruct these in Maya or other modeling software. To afford an effective interaction with serially sectioned reconstructed models, we propose VRKidney. This simulation software acts as VR tools on computer for man-machine interaction. By integrating image registration, reconstruction, VRKidney provides fast method using ITK interactive manipulation VTK. paper discusses the...