Weihua Su

Prof. Dr. Weihua Su

orcid

Research Focus & Metrics

Academic Career Length

12Years (2015-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 August 2024 in Measurement
Verified
Journal Article
Closed access
DZDongyuan Zang
Dongyuan Zang
WSWeihua Su
Weihua Su
BZBonan Zhang
Bonan Zhang
HLHuixin Liu
Huixin Liu

19 September 2024 in Journal of Real-Time Image Processing
Verified
Journal Article
Closed access
PDF file
HLHuixin Liu
Huixin Liu
GLGuohua Lu
Guohua Lu
MLMingxi Li
Mingxi Li
WSWeihua Su
Weihua Su
ZLZiyi Liu
Ziyi Liu
DZDongyuan Zang
Dongyuan Zang

24 January 2024 in Bioengineering
Verified
Journal Article
Open access
PDF file
JPJiahe Peng
Jiahe Peng
WSWeihua Su
Weihua Su
HCHaiyong Chen
Haiyong Chen
JSJingsheng Sun
Jingsheng Sun
ZTZandong Tian
Zandong Tian
Video-based peripheral oxygen saturation (SpO2) estimation, utilizing solely RGB cameras, offers a non-contact approach to measuring blood oxygen levels. Previous studies set a stable and unchanging environment as the premise for non-contact blood oxygen estimation. Additionally, they utilized a small amount of labeled data for system training and learning. However, it is challenging to train optimal model parameters with a small dataset. The accuracy of blood oxygen detection is easily affected by ambient light and subject movement. To address these issues, this paper proposes a contrastive learning spatiotemporal attention network (CL-SPO2Net), an innovative semi-supervised network for video-based SpO2 estimation. Spatiotemporal similarities in remote photoplethysmography (rPPG) signals were found in video segments containing facial or hand regions. Subsequently, integrating deep neural networks with machine learning expertise enabled the estimation of SpO2. The method had good feasibility in the case of small-scale labeled datasets, with the mean absolute error between the camera and the reference pulse oximeter of 0.85% in the stable environment, 1.13% with lighting fluctuations, and 1.20% in the facial rotation situation.

21 May 2015 in BioMedical Engineering OnLine
Verified
Journal Article
Open access
PDF file
QWQiang-Song Wang
Qiang-Song Wang
XZXin-Chang Zhang
Xin-Chang Zhang
RLRuixin Li
Ruixin Li
JSJing-Gong Sun
Jing-Gong Sun
WSWeihua Su
Weihua Su
XZXi-Zheng Zhang
Xi-Zheng Zhang
The combination of drugs and exercise was the effective treatment in bone injure and rebuilding in clinic. As mechanical strain has potential in inducing the differentiation of osteoblasts in our previous study, the further research to investigate the combination of mechanical strain and icariin stimulation on inducing osteoblast proliferation, differentiation and the possible mechanism in MC3T3-E1 cell line. A whole cell enzyme-linked immunosorbent assay that detects the bromodeoxyuridine incorporation during DNA synthesis was applied to evaluate the proliferation. The mRNA expression of alkaline phosphatase (ALP), osteocalcin (OCN), type I collagen (Col I), bone morphogenetic protein-2 (BMP-2) and BMP-4 was detected by real-time reverse-transcription polymerase chain reaction. The activity of ALP was analyzed by ELISA and the protein expression of OCN, Col I and BMP-2 was assessed by western blot. Moreover, the activity of nuclear transcription factor kappa-B (NF-κB) signaling pathway was investigated with the expression of inhibitor of κB (IκB) α, phosphorylation of IκB-α (P-IκB-α), p65, P-p65 by western blot. We observed that compared to single mechanical strain or icariin stimulation, the mRNA and protein expressions of ALP (P < 0.05 or P < 0.01), OCN (P < 0.01) and Col I (P < 0.05 or P < 0.01) were increased significantly by the combination of mechanical strain and icariin stimulation. Moreover, the combination of mechanical strain and icariin stimulation could up-regulate the expression of BMP-2 (P < 0.01) and BMP-4 compared to single mechanical strain or icariin stimulation. The combination of mechanical strain and icariin stimulation could activate NF-κB signaling pathway by increasing the expression of IκB α, P-IκB-α, p65, P-p65 (P < 0.01). The combination of mechanical strain and icariin stimulation could activate the NF-κB pathway to improve the proliferation, differentiation of osteoblast-like cells.

1 January 2016 in BioMedical Engineering OnLine
Verified
Journal Article
Open access
PDF file
RLRuixin Li
Ruixin Li
LDLi Dong
Li Dong
ZBZhao Bin
Zhao Bin
LHLi Hao
Li Hao
LXLeng Xue
Leng Xue
ZXZhang Xizheng
Zhang Xizheng
In the process of bone defective reparation and engineered bone tissue construction, osteoblasts are adhered to the surface of the scaffold materials and impart the external mechanical load to the osteoblasts. So, the dynamic mechanical property of the scaffolds play an important role in the bone tissue repair and it is valuable to research. Material type and the architectural design of scaffolds are also important to facilitate cell and tissue growth. The aim of this study was to prepare a kind of material with good pore connectivity and analyze its dynamic mechanical property. Fabrication and characterization of micro-hydroxyapatite(m-HA)/chitosan(CS) polymer composite scaffolds with well interconnected spherical pore architectures were reports. Micro-HA was prepared by being calcined and ball milled. Paraffin spheres in the range of 160–330 µm were fabricated with a dispersion method and used as the porogen in the fabrication of the scaffolds. Polymer scaffolds were fabricated by the technique of compression molding and particulate leaching method. The effects of the porogen content on the properties of the scaffolds were studied. With the increase of porogen, the pore of the scaffolds increased and became interconnected. Cyclic loading of three scaffolds were tested with 10 % strain under four levels of loading frequency, 0.1, 0.5, 1 and 1.5 Hz. The porous composite scaffolds exhibited a viscosity-elastic behaviour with a maximum stress of 3–4 kPa. At each frequency, modulus value is decreased with the paraffin microspheres content, but there was no significance difference in the peak stress of the three samples. All the samples tested displayed clear hysteresis loops. There was no significance difference in the peak hysteresis of the three samples, and the hysteresis difference values between the sixth compression cycle and the initial cycle for three samples was similar, with no statistically significant differences. Micro-HA/CS composite scaffolds with interconnected spherical macropores were fabricated using pherical paraffin as porogen. The porous composite scaffolds exhibited a viscosity-elastic behaviour with good repeatability. It is benefit to study the influence of the mechanical load on the cell of the scaffold.

22 August 2023 in Electronics
Verified
Journal Article
Open access
PDF file
SZSong Zijing
Song Zijing
WSWeihua Su
Weihua Su
HCHaiyong Chen
Haiyong Chen
MFMianshi Feng
Mianshi Feng
JPJiahe Peng
Jiahe Peng
AZAifang Zhang
Aifang Zhang
Simultaneous localization and mapping (SLAM) is one of the core technologies for intelligent mobile robots. However, when robots perform VSLAM in dynamic scenes, dynamic objects can reduce the accuracy of mapping and localization. If deep learning-based semantic information is introduced into the SLAM system to eliminate the influence of dynamic objects, it will require high computing costs. To address this issue, this paper proposes a method called YF-SLAM, which is based on a lightweight object detection network called YOLO-Fastest and tightly coupled with depth geometry to remove dynamic feature points. This method can quickly identify the dynamic target area in a dynamic scene and then use depth geometry constraints to filter out dynamic feature points, thereby optimizing the VSLAM positioning performance while ensuring real-time and efficient operation of the system. This paper evaluates the proposed method on the publicly available TUM dataset and a self-made indoor dataset. Compared with ORB-SLAM2, the root-mean-square error of the Absolute Trajectory Error (ATE) can be reduced by 98.27%. The system successfully locates and constructs an accurate environmental map in a real indoor dynamic environment using a mobile robot. It is a VSLAM system that can run in real-time on low-power embedded platforms.