XLXiuchang Li

Xiuchang Li

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Xiuchang Li has a total of 21 co-authors. Here are the top co-authors by number of shared publications:

  • Qing-Xue Zhang

    Department of Echocardiography, The Second Affiliated Hospital of Shandong First Medical University , Taian 271000

    2 shared publications
  • Xiao-Hong Liu

    Department of Cardiology, The Second Affiliated Hospital of Shandong First Medical University , Taian 271000

    2 shared publications
  • Xiao-Hui Zhao

    Department of Cardiology, The Second Affiliated Hospital of Shandong First Medical University , Taian 271000

    2 shared publications
  • Zhi-Jian Liu

    Department of Cardiology, The Second Affiliated Hospital of Shandong First Medical University , Taian 271000

    2 shared publications
  • Bin Leng

    Department of Echocardiography, The Second Affiliated Hospital of Shandong First Medical University, Taian, 271000, China

    1 shared publication
  • Guang Zhang

    Department of Health Management, Shandong Engineering Research Center of Health Management, The First Affiliated Hospital of Shandong First Medical University & Shandong Provincial Qianfoshan Hospital, Shandong Institute of Health Management, Jinan, People’s Republic of China

    1 shared publication

Latest Publications

  • Open Access

    Domain adaptation, self-supervision, and generative augmentation enhance GNNs for breast cancer prediction

    by Shi Qiu, Yun Zhao, Xiuchang Li

    Breast cancer presents substantial molecular heterogeneity, requiring accurate subtype classification, receptor-status prediction, and survival estimation for precision care. Existing machine-learning models often fail to generalize across cohorts or adapt to rare subtypes. We propose a unified graph neural network (GNN) framework that integrates multi-task learning, domain-adversarial adaptation, contrastive self-supervision, few-shot meta-learning, and generative augmentation. Gene-expression data from TCGA-BRCA (1084 samples) and METABRIC (1980 samples) were mapped onto gene-centric PPI graphs and patient-similarity graphs. A shared encoder (including Graph Transformer variants) jointly predicts intrinsic subtypes (Luminal A, Luminal B, HER2-enriched, Basal-like), ER/PR/HER2 biomarkers, and overall survival (OS) using a Cox proportional hazards head. Validation included fivefold cross-validation and strict TCGA → METABRIC transfer testing. The multi-task Graph Transformer achieved subtype F1 = 0.872, ER/PR/HER2 AUCs of 0.960/0.943/0.918, and C-index = 0.721. Domain adaptation improved external subtype F1 from 0.738 to 0.801. For the HER2-enriched subtype, MAML enabled few-shot prediction with F1 = 0.782, while MolGAN augmentation increased HER2 AUC to 0.935. GNNExplainer highlighted biologically consistent drivers, including ESR1, ERBB2, and PGR, aligning with known hormonal and HER2 signaling mechanisms. This study introduces a comprehensive, interpretable GNN framework that unifies subtyping, biomarker prediction, and survival modeling while improving cross-cohort robustness and rare-subtype adaptation. The combination of multi-task learning, domain adaptation, self-supervision, and generative augmentation demonstrates strong potential for clinically actionable decision support.


  • Open Access

    The association of epicardial adipose tissue thickness and human neutrophil lipocalin with mitral annular calcification

    by Yeheng Xue, Xinyi Wang, Xiaohong Liu, Qingxue Zhang, Zhijian Liu, Bin Leng, Xiuchang Li

  • Open Access

    Association between circulating selenium levels and arterial stiffness: a nationwide cross-sectional study

    by Kunsheng Zhao, Lei Qiao, Wenqi Hu, Wenhai Sui, Xiuchang Li, Guang Zhang

    Background: Atherosclerotic cardiovascular disease (ASCVD) is the leading cause of death worldwide. Arterial stiffness is an important pathological characteristic of ASCVD. Selenium (Se) may have a notable effect of arterial stiffness. However, the relationship between circulating Se (CSe) levels and arterial stiffness remains unclear. Methods: A total of 11 937 participants from the National Health and Nutrition Examination Survey (NHANES) from 2011 to 2018 were enrolled. The relationship between CSe levels, estimated pulse wave velocity (ePWV), and the atherogenic index of plasma (AIP) was evaluated using univariate and multivariate linear regression, dose-response, and mediation analyses. Results: A linear relationship between the CSe index and high AIP ( P for nonlinearity = 0.1) and a U-shaped relationship between the CSe index and high ePWV were found ( P for nonlinearity < 0.001). In fully adjusted models, the relationship between CSe and high AIP was positive (OR = 1.48, 95% CI: 1.20, 1.82; P = 0.001). An inflection point was found in the relationship between the CSe index and high ePWV. CSe levels of < 2.99 μmol/L were negatively related to high ePWV (OR = 0.71; 95% CI: 0.59, 0.86; P < 0.001). In contrast, CSe levels of ≥ 2.99 μmol/L were positively related to high ePWV (OR = 2.66; 95% CI: 1.66, 4.25; P < 0.001). Furthermore, the indirect effect mediated by the sex of CSe on high AIP was prominent (indirect effect = 0.006; 95% CI: 0.003, 0.008; percent mediation = 26.49%) in the fully adjusted models. Conclusions: Our findings suggests a significant relationship between CSe levels and arterial stiffness, the mechanism behind which should be further explored.


  • A novel indicator derived from heart sound energy map obtained by a wearable device can evaluate LV function

    by Zhi-Jian Liu, Xiao-Hong Liu, Qing-Xue Zhang, Xiao-Hui Zhao, Xiu-Chang Li

    Background: Assessing cardiac function is important for early diagnosis, prognosis, and out-hospital follow-up for heart failure patients. It is especially crucial to identify a cost-effective and easily adaptable approach for evaluating cardiac function at home. Objective: Our purpose was to investigate whether some metrics derived from the heart sound energy map obtained by a wearable heart acquisition device could reflect the cardiac function. Methods: 99 subjects were categorized into two groups based on whether their left ventricular ejection fraction was <50%. In the heart sound energy map mode, time intervals of energy intensity between 2100 and 2400 of the first heart sound was denoted as “a,” 1500–1800 as “b,” 1200–1500 as “c,” and 600–900 as “d.” The ratio of the time difference between adjacent energy intensity intervals and the high-energy interval was calculated, referred to as the Red index [(b-a)/a], Brn index [(c-b)/b], and Yel index [(d-c)/c], respectively. Results: Significant differences were observed in the Red index and Yel index between the two groups (P < .001 and P = 0.017, respectively). The Red and Yel indices showed a significant correlation with ejection fraction (EF) value (r = −0.50 and −0.27, P < .001 and P = .008, respectively). The intraclass correlation coefficient conformance test indicated that the intragroup correlation coefficients of the Red index with EF were 0.57. Receiver operating characteristic analysis revealed that the Red index exhibits strong diagnostic value for EF values <50% with an AUC of 0.86 (95% CI: 78% ~ 94%). Conclusions: The Red index, derived from the heart sound energy map, demonstrates a significant correlation with both the EF value and Tei index and can identify the reduction of left ventricular EF.

  • Open Access

    The Correlation between Epicardial Adipose Tissue Thickness Measured by Echocardiography and P-Wave Dispersion and Atrial Fibrillation

    by Qing-Xue Zhang, Zhi-Jian Liu, Xiao-Hong Liu, Xiao-Hui Zhao, Xiu-Chang Li

    Background: Recent studies have indicated a close relationship between the thickness of epicardial adipose tissue (EAT) and the occurrence as well as persistence of atrial fibrillation (AF). However, the pathogenesis of this association is still in the exploratory stage. The aim of this study is to explore the correlation EAT, as measured by echocardiography, and P-wave dispersion (Pd) in the context of atrial fibrillation. Additionally, the study seeks to analyze the utility of EAT at different anatomical sites in identifying individuals who are predisposed to atrial fibrillation. Methods: A total of 136 subjects were enrolled and categorized into groups based on the guidelines: paroxysmal atrial fibrillation group (PAF group), persistent atrial fibrillation group (AF group), and non-atrial fibrillation group. Comprehensive clinical data, including general information and medications that could impact the occurrence of atrial fibrillation, were gathered for all patients. Echocardiography was employed to measure the maximum EAT thickness near the apex of the heart on the anterior right ventricular wall and near the base of the right ventricle for each participant. Pd values were computed for each patient based on standard 12-lead synchronous electrocardiogram (ECG). The study involved comparing the disparity in EAT thickness between the two specified sites across the three groups. Additionally, correlation analyses were performed to assess the relationship between EAT thickness at the two sites and Pd. Regression analysis was applied to explore potential risk factors for atrial fibrillation. The diagnostic value of EAT at each site in predicting atrial fibrillation was evaluated using Receiver Operating Characteristic curve (ROC) analysis. Results: EAT thickness of the anterior wall near the apex of the heart and near the base of the right ventricle were significantly positively correlated with Pd (p < 0.05), EAT thickness near the base and left atrial diameter were independent risk factors for atrial fibrillation (OR = 13.673, 95% CI 2.819~66.316, p = 0.001; OR = 2.294, 95% CI 1.020~5.156, p = 0.045). ROC analysis showed that the area under the curve of EAT thickness near the heart base was 0.723, and the best threshold for predicting the occurrence of AF was 1.05 cm. Conclusions: The echocardiography-measured epicardial adipose tissue thickness, particularly in proximity to the heart base, exhibits a significant correlation with Pd. Notably, EAT thickness near the heart base demonstrates superior predictive capability for atrial fibrillation compared to thickness near the apex.


  • Attenuation of Myocardial Fibrosis Using Molecular Hydrogen by Inhibiting the TGF-β Signaling Pathway in Spontaneous Hypertensive Rats

    by Libo Liu, Qian Shi, Xiaohong Liu, Yanmin Li, Xiuchang Li

    Previous studies have shown that hydrogen can antagonize the fibrosis of various organs. We investigated whether hydrogen-rich saline (HRS) can attenuate myocardial fibrosis in spontaneously hypertensive rats (SHRs) and clarified the mechanisms involved. We examined the effect of HRS and pirfenidone (PFD) on myocardial fibrosis in SHR. Systolic blood pressure, left ventricular mass index (LVMI), and heart weight index (HWI) were measured, Masson trichrome staining was performed. We assessed the role of superoxide dismutase (SOD), malondialdehyde (MDA), Alpha-smooth muscle actin (α-SMA), collagen I, collagen III, and tissue inhibitors of metalloproteinases (TIMPs) in myocardium. We detected the concentrations of procollagen type-I C-terminal propeptide (PICP), procollagen type-III N-terminal propeptide (PIIINP), and angiotensin II (Ang II) in rat serum. Furthermore, the relative protein levels of the transforming growth factor beta (TGF-β)/Smad pathway were tested. We discovered that HRS decreases LVMI (P < 0.05) and HWI (P < 0.05) in vivo. Compared with model group, HRS decreases the level of collagen volume fraction (P < 0.0001), collagen I (P < 0.001), and collagen III (P < 0.001) in myocardium, and Ang II (P < 0.05), PICP (P < 0.001), and PIIINP (P < 0.05) in serum. In addition, HRS downregulates the expression of MDA (P < 0.01), α-SMA (P < 0.05), and TIMPs (P < 0.05), and increased SOD (P < 0.05). Furthermore, HRS downregulated the expression levels of TGF-β1 (P < 0.0001), Smad3 (P < 0.0001), and Smad2/3 (P < 0.001), but had no effect on Smad7 expression (P > 0.05). PFD had similar effect compared with HRS and control group. HRS reduced oxidative stress and improved myocardial collagen content, which may be related to inhibition of the TGF-β signaling pathway. This suggests that HRS is an effective therapeutic strategy for myocardial fibrosis.