A5107117816- Human Pose and Action Recognition
- Retinal Imaging and Analysis
- Advanced Neural Network Applications
- Ion channel regulation and function
- Biochemical Analysis and Sensing Techniques
- COVID-19 diagnosis using AI
- Video Surveillance and Tracking Methods
- Neuroblastoma Research and Treatments
- Digital Imaging for Blood Diseases
- Lung Cancer Research Studies
- Power Systems and Technologies
- Human Motion and Animation
- Cancer therapeutics and mechanisms
- Autonomous Vehicle Technology and Safety
- Cardiac electrophysiology and arrhythmias
- Retinal and Optic Conditions
- Neurotransmitter Receptor Influence on Behavior
- Retinal Diseases and Treatments
- Neuroscience and Neural Engineering
- Power Line Inspection Robots
- Psychedelics and Drug Studies
- Advanced Sensor and Control Systems
- 3D Shape Modeling and Analysis
Fudan University
2022-2024
Shandong Province Meteorological Bureau
2019
Institute of Pharmacology
2013
Psychedelics alter consciousness and may have potential for drug development. As psychedelics are likely therapeutically active, it is important to study their effects mechanisms using preclinical models. Here, we examined the of phenylalkylamine indoleamine on locomotor activity exploratory behaviour mouse Behavioural Pattern Monitor (BPM). DOM, mescaline, psilocin reduced at high doses influenced rearings, an behaviour, in a characteristic inverted U-shaped dose-response function....
Many studies have provided convincing evidence for hERG as an important diagnostic and prognostic factor in human cancers, well a useful target antineoplastic therapy. Our previous study also revealed that knockdown of herg gene expression by shRNA interference inhibited the growth neuroblastoma cells vitro vivo. In experiment, novel 4-amino piperidine analog, ZC88, was examined its effect on potassium channels antitumor potency observed The results showed ZC88 could block hERG1 hERG1b...
Diabetic Retinopathy (DR) is a diabetic complication that predisposes patients to visual impairments could lead blindness. Lesion segmentation using deep learning algorithms an effective measure screen and prevent early DR. However, there are several types of DR with varying sizes high inter-class similarity, making difficult. In this paper, we propose supervised method (MSLF-Net) based on multi-scale–multi-level feature fusion achieve accurate end-to-end lesion segmentation. MSLF-Net builds...
To improve the accuracy of retinal blood vessel segmentation, this paper presents a supervised segmentation method Multi-type Feature Enhancement (MFE). In MFE, high-low level feature fusion module (FFM) is constructed to realize spatial information and semantic information, which provides more comprehensive features for segmentation. Spatial Information Module (SIEM) introduced promote learning ability features, available image with continuous morphology. addition, an operation...
Human instance segmentation in occlusion scenarios remains a challenging task, especially nursing scenarios, which hinders the development of robots. Existing approaches are unable to focus network's attention on occluded areas, leads unsatisfactory results. To address this issue, paper proposes novel and effective network based density map task. Density map-based neural networks perform well cases where human bodies occlude each other can be trained without additional annotation...
Abstract：The wind speed and direction meter of meteorological station is installed on the pole 10 meters high from observation site, which also causes some difficulties for maintenance in later period. Based this situation, an electric lifting worktable (referred to as worktable) specially designed sites. It can lift personnel equipment tools a height convenient work, then grasp through mechanical arm. be operated by single person, thus getting rid traditional dimension climbing or releasing...