Minhua Lu

Minhua Lu

Auto-generated profile, based on publicly available information

Minhua Lu

Shenzhen University

Research Focus & Metrics

Academic Career Length

36Years (1991-Present)

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

Representative Publications

View All Publications
26 December 2022 in IEEE Transactions on Biomedical Engineering
Verified
Journal Article
Closed access
HZHao Zhu
Hao Zhu
JWJun Wang
Jun Wang
YZYin-Ping Zhao
Yin-Ping Zhao
MLMinhua Lu
Minhua Lu
JSJun Shi
Jun Shi
The resting-state functional magnetic resonance imaging (rs-fMRI) faithfully reflects the brain activities and thus provides a promising tool for autism spectrum disorder (ASD) classification. Up to now, graph convolutional networks (GCNs) have been successfully applied in rs-fMRI based ASD classification. However, most of these methods were developed based on functional connectivities (FCs) that only reflect low-level correlation between brain regions, without integrating both high-level discriminative knowledge and phenotypic information into classification. Besides, they suffered from the overfitting problem caused by insufficient training samples. To this end, we propose a novel contrastive multi-view composite GCN (CMV-CGCN) for ASD classification using both FCs and HOFCs. Specifically, a pair of graphs are constructed based on the FC and HOFC features of the subjects, respectively, and they share the phenotypic information in the graph edges. A novel contrastive multi-view learning method is proposed based on the consistent representation of both views. A contribution learning mechanism is further incorporated, encouraging the FC and HOFC features of different subjects to have various contribution in the contrastive multi-view learning. The proposed CMV-CGCN is evaluated on 613 subjects (including 286 ASD patients and 327 NCs) from the Autism Brain Imaging Data Exchange (ABIDE). We demonstrate the performance of the method for ASD classification, which yields an accuracy of 75.20% and an area under the curve (AUC) of 0.7338. Experimental results show that our proposed method outperforms state-of-the-art methods on the ABIDE database.

24 July 2015 in BioMedical Engineering OnLine
Verified
Journal Article
Open access
PDF file
YCYujie Chen
Yujie Chen
HXHuahua Xiong
Huahua Xiong
Dan WuDan Wu
Dan Wu
Sandeep PirbhulalSandeep Pirbhulal
Sandeep Pirbhulal
XTXiaohong Tian
Xiaohong Tian
HWHuang Wenhua
Huang Wenhua
High blood pressure (BP) is among significant risk factor for stroke and other vascular occurrences, it experiences nonstop fluctuations over time as a result of a complex interface among cardiovascular control mechanisms. Large blood pressure variability (BPV) has been proved to be promising in providing potential regulatory mechanisms of the cardiovascular system. Although the previous studies also showed that BPV is associated with increased carotid intima-media thickness (IMT) and plaque, whether the correlation between variability in blood pressure and left common carotid artery-intima-media thickness (LCCA-IMT) is stronger than right common carotid artery-intima-media thickness (RCCA-IMT) remains uncertain in hypertension. We conduct a study (78 hypertensive subjects, aged 28–79) to evaluate the relationship between BPV and carotid intima-media thickness in Shenzhen. The blood pressure was collected using the 24 h ambulatory blood pressure monitoring, and its variability was evaluated using standard deviation (SD), coefficient of variation (CV), and average real variability (ARV) during 24 h, daytime and nighttime. All the IMT measurements are collected by ultrasound. As the results showed, 24 h systolic blood pressure variability (SBPV) evaluated by SD and ARV were significantly related to LCCA-IMT (r1 = 0.261, P = 0.021; r1 = 0.262, P = 0.021, resp.). For the daytime diastolic blood pressure variability (DBPV), ARV indices were significantly related to LCCA-IMT (r1 = 0.239, P = 0.035), which differed form BPV evaluated by SD and CV. For the night time, there is no significant correlation between the BPV and IMT. Moreover, for all the subjects, there is no significant correlation between the BPV and RCCA-IMT/number of plaques, whereas, the SD, CV, and ARV of daytime SBP showed a positive correlation with LCCA-IMT (r1 = 0.312, P = 0.005; r1 = 0.255, P = 0.024; r1 = 0.284, P = 0.012, resp.). Moreover, the ARV of daytime SBPV, 24 h SBPV and nighttime DBPV showed a positive correlation with the number of plaques of LCCA (r1 = 0.356, P = 0.008; r1 = 0.297, P = 0.027; r1 = 0.278, P = 0.040, resp.). In addition, the number of plaques in LCCA had higher correlation with pulse pressure and diastolic blood pressure than that in RCCA. And multiple regression analysis indicated LCCA-IMT might not only be influenced by age or smoking but also by the SD index of daytime SBPV (p = 0.035). The results show that SBPV during daytime and 24 h had significant correlation with IMT, for the hypertensive subjects from the southern area of China. Moreover, we also found the daytime SBPV to be the best predictor for the progression of IMT in multivariate regression analysis. In addition, the present study suggests that the correlation between BPV and left common carotid artery—intima-media thickness/number of plaques is stronger than right common carotid artery-intima-media thickness/number of plaques.

3 February 2016 in International Journal of Machine Learning and Cybernetics
Verified
Journal Article
Closed access
PDF file
RMRui Mao
Rui Mao
PZPeihan Zhang
Peihan Zhang
XLXingliang Li
Xingliang Li
XLXi Liu
Xi Liu
MLMinhua Lu
Minhua Lu

20 November 2014 in BioMedical Engineering OnLine
Verified
Journal Article
Open access
PDF file
HXHuahua Xiong
Huahua Xiong
XLXin Liu
Xin Liu
XTXiaohong Tian
Xiaohong Tian
LPLina Pu
Lina Pu
HZHeye Zhang
Heye Zhang
YZYuan-Ting Zhang
Yuan-Ting Zhang
Blood pressure (BP) is associated with early atherosclerosis and plaque rupture because the BP variability can significantly affect the blood flow velocity and shear stress over the plaque. However, the mechanical response of BP variability to the plaque remains unclear. Therefore, we investigated the correlation between different maximum systolic blood pressure (SBP) and the stress distribution on plaque, as well as the stress over the plaque and blood velocity around the plaque using different BP variations, which are the BP variability in different phases during one cardiac cycle and beat-to-beat BP variability. We established a two-dimensional artery model with stenosis at the degree of 62.5%. Eight combinations of pulsatile pressure gradients between the inflow and outflow were implemented at the model. Three levels of fibrous cap thickness were taken into consideration to investigate the additional effect on the BP variability. Wall shear stress and stress/strain distribution over the plaque were derived as well as the oscillation shear index (OSI) to analyze the impact of the changing rate of BP. The stresses at diastole were 2.5% ± 1.8% lower than that at systole under the same pressure drop during one cycle. It was also found that elevated SBP might cause the immediate increment of stress in the present cycle (292% ± 72.3%), but slight reduction in the successive cycle (0.48% ± 0.4%). The stress/strain distribution over the plaque is sensitive to the BP variability during one cardiac cycle, and the beat-to-beat BP variability could cause considerable impact on the progression of atherosclerosis in long-term.

16 September 2014 in Neurocomputing
Verified
Journal Article
Closed access
JSJun Shi
Jun Shi
QJQikun Jiang
Qikun Jiang
RMRui Mao
Rui Mao
MLMinhua Lu
Minhua Lu
TWTianfu Wang
Tianfu Wang

1 January 2021 in Computerized Medical Imaging and Graphics
Verified
Journal Article
Closed access
CXChenhui Xiao
Chenhui Xiao
ZLZhenzhou Li
Zhenzhou Li
JLJianfeng Lu
Jianfeng Lu
JWJinyan Wang
Jinyan Wang
HZHaoteng Zheng
Haoteng Zheng
MLMinhua Lu
Minhua Lu