- Muscle activation and electromyography studies
- Lower Extremity Biomechanics and Pathologies
- Infrared Thermography in Medicine
- Diabetic Foot Ulcer Assessment and Management
- Ultrasound Imaging and Elastography
- Body Composition Measurement Techniques
- Stroke Rehabilitation and Recovery
- Ocular Surface and Contact Lens
- Photoacoustic and Ultrasonic Imaging
- Glaucoma and retinal disorders
- Corneal surgery and disorders
- Radiomics and Machine Learning in Medical Imaging
- Hand Gesture Recognition Systems
- Advanced Sensor and Energy Harvesting Materials
- Retinal Imaging and Analysis
- Medical Imaging and Analysis
- AI in cancer detection
- Sports Performance and Training
- Image and Object Detection Techniques
- Sports injuries and prevention
- Liver Disease Diagnosis and Treatment
- Optical measurement and interference techniques
- Advanced MRI Techniques and Applications
- EEG and Brain-Computer Interfaces
- Non-Invasive Vital Sign Monitoring
Shenzhen University
2016-2025
Second Affiliated Hospital & Yuying Children's Hospital of Wenzhou Medical University
2016-2025
Wenzhou Medical University
2016-2025
Shenzhen University Health Science Center
2016-2025
Guiyang Medical University
2025
Hong Kong Polytechnic University
2006-2023
Key Laboratory of Guangdong Province
2015-2022
Lishui University
2022
Zhejiang University
2022
East Asia School of Theology
2021
Assessing the location and extent of lesions caused by chronic stroke is critical for medical diagnosis, surgical planning, prognosis. In recent years, with rapid development 2D 3D convolutional neural networks (CNN), encoder-decoder structure has shown great potential in field image segmentation. However, CNN ignores information images, while suffers from high computational resource demands. This paper proposes a new architecture called dimension-fusion-UNet (D-UNet), which combines...
Flexible strain sensors that can be comfortably attached onto the skin for real-time and accurate detection of physiological signals are particularly important realization healthcare, soft robotics human-machine interfacing. It is still challenging to develop satisfy both high stretchability sensitivity. Here, we demonstrate an ultra-stretchable, highly sensitive biocompatible capacitive-type sensor based on a nanocomposite containing ionic hydrogels silver nanofibers (AgNFs). The exhibits...
This paper proposes a segmentation-free radiomics method to classify malignant and benign breast tumors with shear-wave elastography (SWE) data. The is targeted integrate the advantage of both SWE in providing important elastic morphology information convolutional neural network (CNN) automatic feature extraction accurate classification. Compared traditional methods, proposed designed directly extract features from dataset without prerequisite segmentation manual operation. can keep...
Breast density is widely adopted to reflect the likelihood of early breast cancer development. Existing methods mammographic classification either require steps manual operations or achieve only moderate accuracy due limited model capacity. In this study, we present a radiomics approach based on dilated and attention-guided residual learning for task classification. The proposed method was instantiated with two datasets, one clinical dataset publicly available dataset, accuracies 88.7 70.0...
Microvascular invasion (MVI) has a significant effect on the prognosis of hepatocellular carcinoma (HCC), but its preoperative identification is challenging. Radiomics features extracted from medical images, such as magnetic resonance (MR) can be used to predict MVI. In this study, we explored effects different imaging sequences, feature extraction and selection methods, classifiers performance HCC MVI predictive models.After screening against inclusion criteria, 69 patients with gadoxetic...
Myopia control interventions, such as defocus incorporated multiple segments (DIMS) spectacle lenses, have been adopted in school-aged children to reduce the prevalence of myopia and its complications. This study aimed investigate effect DIMS lenses on subfoveal choroidal thickness (SfChT) over a period two years, response myopic is crucial factor exploring potential predicting progression.This involved secondary analysis our previous randomized clinical trial. Myopic aged 8-13 years were...
A novel approach for personal identification using finger images which exploits the orientation features from random knuckle lines finite Radon transform is proposed. The feasibility of this rigorously evaluated on a publically available database 158 subjects and achieves highly promising results.
Muscle fascicle pennation angle (PA) is an important parameter related to musculoskeletal functions, and ultrasound imaging has been widely used for measuring PA, but manually frame by in most cases. We have earlier reported automatic method estimate aponeurosis orientation based on Gabor transform Revoting Hough Transform (RVHT). In this paper, we proposed a the overall of muscle fascicles region interest, order complete computing other side angle, found RVHT. The measurements orientations...
Transient elastography (TE), recommended by the WHO, is an established method for characterizing liver fibrosis via stiffness measurement (LSM). However, technical barriers remain towards point-of-care application, as conventional TE requires wired connections, possesses a bulky size, and lacks adequate imaging guidance precise localization. In this work, we report design, phantom validation, clinical evaluation of palm-sized system that enables simultaneous B-mode LSM. The performance was...
Abstract An accurate segmentation and quantification of the superficial foveal avascular zone (sFAZ) is important to facilitate diagnosis treatment many retinal diseases, such as diabetic retinopathy vein occlusion. We proposed a method based on deep learning for automatic sFAZ in optical coherence tomography angiography (OCTA) images with robustness brightness contrast (B/C) variations. A dataset 405 OCTA from 45 participants was acquired Zeiss Cirrus HD-OCT 5000 ground truth (GT) manually...
Recently, an emerging trend in medical image classification is to combine radiomics framework with deep learning network integrated system. Although this combination efficient some tasks, the learning-based often difficult capture effective representation of lesion regions, and prone face challenge overfitting, leading unreliable features inaccurate results, especially when sizes lesions are small or training dataset small. In addition, these combinations mostly lack feature selection...
18F-fluorodeoxyglucose (FDG)-positron emission tomography (PET) is an essential component in the diagnostic work-up of patients with Parkinson's Disease (PD). With development pattern recognition technology, analysis brain images using deep learning are becoming more and popular. However, existing computer-aided-diagnosis technologies often overfit have poor generalizability. Therefore, we aimed to improve a framework based on Group Lasso Sparse Deep Belief Network (GLS-DBN) for...
To develop a nonenhanced CT-based radiomic signature for the differentiation of iodinated contrast extravasation from intraparenchymal haemorrhage (IPH) following mechanical thrombectomy.Patients diagnosed with acute ischaemic stroke who underwent thrombectomy in 4 institutions December 2017 to June 2020 were included this retrospective study. The study population was divided into training cohort and validation cohort. CT images taken after used extract features. maximum relevance minimum...
Improvement in muscle strength is an important aim for the rehabilitation of hemiplegia patients. Presently, prescription depends on evaluation results strength, which are routinely estimated by experienced physicians and therefore not finely quantitative. Widely-used quantification methods disability, such as Barthel Index (BI) motor component Functional Independent Measure (M-FIM), yet have limitations their application, since both them differentiated disability better lower than higher...
Unilateral peripheral facial paralysis (UPFP) is a form of nerve and clinically classified according to conditions symmetry. Prompt precise assessment crucial neural rehabilitation UPFP. The prevalent House-Brackmann (HB) grading system relies on subjective judgments with significant interobservation variation. Therefore, explore an objective method for the UPFP assessment, clinical image sequences are captured using web camera setup while 5 healthy 27 subjects perform group predefined...
Unilateral peripheral facial paralysis (UPFP) is a form of nerve and clinically classified according to asymmetry. Prompt precise assessment crucial the neural rehabilitation UPFP. For UPFP assessment, most existing systems are subjective empirical. Therefore, an objective system will help clinical doctors obtain prompt assessment. Distinguishing precisely between degrees asymmetry hard using pure pattern recognition methods. Thus, novel process based on convolutional neuronal networks...
Importance Sleep disorders and mild cognitive impairment (MCI) commonly coexist in older adults, increasing their risk of developing dementia. Long-term tai chi chuan has been proven to improve sleep quality adults. However, adherence extended training regimens can be challenging. Repetitive transcranial magnetic stimulation (rTMS) is a neuromodulation technique that may enhance the benefits exercise. Objective To investigate whether 1-Hz rTMS right dorsolateral prefrontal cortex could...
Objective: To investigate the influence of medical environment familiarity on preoperative anxiety, anesthesia induction compliance, and emergence agitation (EA) in children undergoing adenotonsillectomy a day surgery setting. Methods: A retrospective analysis was conducted elective setting under general at Second Affiliated Hospital Wenzhou Medical University between January 2018 December 2019, October 2020 November 2022, March 2023. The were divided into three groups according to their...
BACKGROUND Multiple primary malignant tumors refer to the occurrence of two or more in same organ multiple organs tissues at time successively patient, and can occur anywhere body. The treatment guidelines for patients with are currently controversial. CASE SUMMARY A 51-year-old male patient liver cancer portal hypertension received 42 months co-treatment atezolizumab bevacizumab. After that, disease was rated stable disease. then diagnosed gastric cancer. Since not sensitive anti-programmed...