- Human Pose and Action Recognition
- Gait Recognition and Analysis
- Hand Gesture Recognition Systems
- Cancer Diagnosis and Treatment
- Head and Neck Cancer Studies
- Video Surveillance and Tracking Methods
- Ectopic Pregnancy Diagnosis and Management
- Uterine Myomas and Treatments
- Maternal and fetal healthcare
- Image Enhancement Techniques
- Pregnancy-related medical research
- Pelvic floor disorders treatments
- Stroke Rehabilitation and Recovery
- Medical Imaging and Pathology Studies
- Multimodal Machine Learning Applications
- Impact of Light on Environment and Health
- Maternal and Perinatal Health Interventions
- Advanced Image Fusion Techniques
- Anesthesia and Pain Management
- Image Processing Techniques and Applications
- Visual Attention and Saliency Detection
- Image and Signal Denoising Methods
- Cardiac Health and Mental Health
- Gynecological conditions and treatments
- Ureteral procedures and complications
Wenzhou Medical University
2012-2025
First Affiliated Hospital of Wenzhou Medical University
2019-2025
National University of Singapore
2023-2025
Women's Hospital, School of Medicine, Zhejiang University
2021-2023
NetEase (China)
2023
Changsha Hospital for Maternal and Child Health Care
2015-2022
Beijing Jiaotong University
2021-2022
Weatherford College
2021
Credo Reference (United States)
2021
Gait recognition is one of the most important biometric technologies and has been applied in many fields. Recent gait frameworks represent each frame by descriptors extracted from either global appearances or local regions humans. However, representations based on information often neglect details frame, while region cannot capture relations among neighboring regions, thus reducing their discriminativeness. In this paper, we propose a novel feature extraction fusion framework to achieve...
Visibility in hazy nighttime scenes is frequently reduced by multiple factors, including low light, intense glow, light scattering, and the presence of multicolored sources. Existing dehazing methods often struggle with handling glow or low-light conditions, resulting either excessively dark visuals unsuppressed outputs. In this paper, we enhance visibility from a single haze image suppressing enhancing regions. To handle effects, our framework learns rendered pairs. Specifically, source...
Gait recognition is a biometric technology that recognizes the identity of humans through their walking patterns. Compared with other technologies, gait more difficult to disguise and can be applied condition long-distance without cooperation subjects. Thus, it has unique potential wide application for crime prevention social security. At present, most methods directly extract features from video frames establish representations. However, these architectures learn representations different...
Gait benchmarks empower the research community to train and evaluate high-performance gait recognition systems. Even though growing efforts have been devoted cross-view recognition, academia is restricted by current existing databases captured in controlled environment. In this paper, we contribute a new benchmark strong baseline for REcognition Wild (GREW). The GREW dataset constructed from natural videos, which contain hundreds of cameras thousands hours streams open With tremendous manual...
Post-stroke depression (PSD) constitutes an important complication of stroke, affecting approximately one-third stroke patients. PSD decreases rehabilitation motivation, delays function recovery, and increases the family social burden Motivational interviewing (MI) may be effective practical intervention strategy, but its effectiveness in improving remains uncertain. A parallel two-group quasi-experimental study was conducted. Patients with early were recruited from neurology department a...
Most gait recognition methods exploit spatial-temporal representations from static appearances and dynamic walking patterns. However, we observe that many part-based neglect at boundaries. In addition, the phenomenon of overfitting on training data is relatively common in recognition, which perhaps due to insufficient low-informative silhouettes. Motivated by these observations, propose a novel mask-based regularization method named ReverseMask. By injecting perturbation feature map,...
To determine the biometry of anterior segment dimensions human eye on both horizontal and vertical meridians with extended scan depth optical coherence tomography (OCT) during accommodation.Twenty pre-presbyopic volunteers, aged between 24 30, were recruited. The ocular each subject was imaged using an OCT under non- 3.0 diopters (D) accommodative demands meridians. All images analyzed to yield following parameters: pupil diameter (PD), chamber (ACD), posterior surface curvatures crystalline...
Due to the lack of health education adherence assessment tools for stroke patients, in this population is insufficient, which hinders prevention and rehabilitation stroke. This study aims develop validate a Health Education Adherence Scale Stroke Patients (HEAS-SP).A cross-sectional design with purposive sampling method was used study. Six hundred fifty-four eligible participants completed demographic questionnaire HEAS-SP. The data collection lasted 7 months, from March 1stto September 30th...
Pain is one of the most common and intolerable symptoms in cancer patients. But pain control still negative China.This paper explores application quality circle treatment inpatients to improve life patients with pain.Established a group analyze current status moderate pain, set goals, formulate corresponding countermeasures implement review them stages. The plan-do-check-act method was cyclically applied. Brief Inventory used evaluate hospitalized before (January April 2016) after (September...
Generic visual object tracking is challenging due to various difficulties, e.g. scale variations and deformations. To solve those problems, we propose a novel multi-scale selective kernel module for tracking, which contains small-scale large-scale branches model the target at different scales attention mechanism capture more effective appearance information of target. In our module, cascade multiple convolutional blocks as an equivalent branch extract features effectively. Besides, present...
Abstract: Metastases from colorectal cancer can occur either through the lymphatic or by hematogenous spread. The most common metastatic sites are lung and liver. Nasopharyngeal metastasis has never been previously reported in literature on internet databases we found. In this paper, present case of a 79-year-old male suffering adenocarcinoma rectum with distant metastases to liver, lung, nasopharynx. Over previous 7 years, he had received treatment for rectal including radical surgery...
Objectives To examine the association between illness uncertainty, resourcefulness, and poststroke depression (PSD) identifies whether stroke patients’ resourcefulness plays a mediating role in relationship uncertainty PSD. Methods A cross-sectional study was conducted from September 2020 to April 2021. convenience sample of 355 patients recruited. general characteristic questionnaire, Mishel Uncertainty Illness Scale, Resourcefulness Scale (RS), Patient Health Questionnaire-9 (PHQ-9) were...
In the problem of image deblurring, restoration details in severely blurred images has always been difficult. this paper, we focus on effectively eliminating ringing artifact and wrinkles that appear after deburring, propose a novel blind debluring method based dual attention deep prior (DADIP) network 2-dimensional (2D) blur kernel estimation with convolutional neural (CNN). DADIP network, mechanism is firstly combined squeeze excitation (SENet), which greatly improves effect details. More...
Abstract Aim To validate risk factors and a nomogram prediction model for the failure of trial labor after cesarean section (TOLAC) in Chinese population. Methods We included women who tried TOLAC between January 2017 May 2019, grouped according to success/failure TOLAC. The patients were randomized 3:1 into development validation sets. Multivariable logistic regression analyses used develop failure. Results In total, 535 (86.3%) ( n = 620) aged 29–34 years had successful vaginal birth...
Gait recognition is one of the most important biometric technologies and has been applied in many fields. Recent gait frameworks represent each frame by descriptors extracted from either global appearances or local regions humans. However, representations based on information often neglect details frame, while region cannot capture relations among neighboring regions, thus reducing their discriminativeness. In this paper, we propose a novel feature extraction fusion framework to achieve...
目的 探讨胸腔内注入沙培林对婴儿乳糜胸的临床疗效。 方法 回顾性分析1例婴儿乳糜胸的临床治疗过程,并复习文献,探讨沙培林对婴儿乳糜胸的治疗效果。 结果 患儿,男,61 d,以咳嗽、痰鸣1周,加重伴气促2 d为主要表现。查体:呼吸稍促,左肺呼吸音偏低,左下肺叩诊呈浊音。辅助检查:X线胸片示左侧胸腔大量积液,胸水乳糜定性试验阳性。入院后考虑先天性乳糜胸,先后予持续胸腔闭式引流2周,经胸腔内注入红霉素2个疗程,胸导管结扎术,胸腔引流液均未减少。最后通过胸腔内注入沙培林2次后胸腔积液消失。随访1年无再发胸腔积液。 结论 沙培林能快速、有效地治愈婴儿乳糜胸,是治疗儿童乳糜胸可供选择的一种硬化剂。
Objective To investigate the outcomes of conventional pelvic floor rehabilitation combined with electrical stimulation and biofeedback for treatment postpartum muscle dysfunction. Methods A total 198 women diagnosed dysfunction after natural labor at our hospital between January 2014 June were included in this study. According to personal intention, patients divided into group (n=99) control (n=99). On day 42 postpartum, received (pelvic exercise) , whereas addition rehabilitation....
Abstract Background The levonorgestrel-releasing intrauterine system (LNG-IUS) is an efficient contraceptive and nonsurgical treatment for gynecological conditions. However, its expulsion rates are concerning. In this study, we suggest the use of non-absorbable suture fixation through hysteroscopy to mitigate expulsions evaluate safety outcomes procedure. Methods This study included 73 women (average age: 41.34 years) with a history LNG-IUS who underwent hysteroscopic at Women’s Hospital,...
Background Glucocorticoid (GC) excess is associated with metabolic syndrome and hypogonadism. 11β‐hydroxysteroid dehydrogenase 1 (11βHSD1) regulates intracellular GC concentration thus being a therapeutic target. Methods A new curcumin analogue, (1E,4E)‐1,5‐bis(2‐bromophenyl)penta‐1,4‐dien‐3‐one (GL63) was synthesized, we compared the effect of GL63 in 11β‐HSD1 oxidase activity reductase primary rat Leydig cells, CHOP cell lines transfected human 11βHSD1 gene vitro , vivo effects...