FZFang Zhang

Fang Zhang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Fang Zhang has a total of 12 co-authors. Here are the top co-authors by number of shared publications:

  • Shiben Zhu

    Department of Infectious Diseases, Nanfang Hospital, Southern Medical University, Guangzhou, China

    5 shared publications
  • Qihang Chen

    School of Nursing and Health Studies, Hong Kong Metropolitan University, Hong Kong, Hong Kong SAR, China

    3 shared publications
  • Wenyan Wu

    Medical Laboratory of Shenzhen Luohu People’s Hospital, Shenzhen, Guangdong, China

    3 shared publications
  • Xiaoliu Liu

    Medical Laboratory of Shenzhen Luohu People’s Hospital, Shenzhen, Guangdong, China

    3 shared publications
  • Ziyu Hao

    Jockey Club School of Public Health and Primary Care, The Chinese University of Hong Kong, Hong Kong, Hong Kong SAR, China

    3 shared publications
  • Jiani Yan

    Department of Science and Education, Shenzhen Baoan Women's and Children's Hospital, Shenzhen, China

    2 shared publications

Latest Publications

  • Open Access

    A two-sample bidirectional Mendelian randomization analysis between telomere length and hyperthyroidism

    by Shiben Zhu, Ziyu Hao, Qihang Chen, Xiaoliu Liu, Wenyan Wu, Fang Zhang

    Background: hyperthyroidism characterized by low thyrotropin, highlighting complications and risks, including cardiac issues, osteoporosis, adverse pregnancy outcomes, unintentional weight loss, and increased mortality associated with untreated hyperthyroidism. However, the casual association between telomere length (TL) and hyperthyroidism remains unclear.Objective: We aim to explore the casual relationship between TL and hyperthyroidism.Methods: A two-sample bidirectional Mendelian randomization (MR) analysis employed the inverse variance weighted (IVW) method, supplemented by additional approaches such as Weighted Median (WM), and MR Egger.Results: The summary statistics for TL were derived from the UK Biobank, comprising 472,174 individuals, while the data for hyperthyroidism were sourced from the GWAS Catalog and the FinnGen database, encompassing cohorts of 460,499 and 173,938 individuals, respectively. Utilizing 139 genome-wide significant single nucleotide polymorphisms (SNPs) as instrumental variables (IVs) for TL, forward MR analyses indicated a negative causal effect of TL on hyperthyroidism. The risk of hyperthyroidism decreased as genetically predicted TL increased by one standard deviation, as determined by the IVW form GWAS Catalog (OR:0.659,95%CI: 0.541-0.802, p <0.001) and IVW from FinnGen(OR:0.634, 95%CI: 0.479-0.840, p = 0.001). Other MR methods exhibited a consistent trend in the impact of TL on hyperthyroidism. Reverse MR analysis suggested no causal association between TL and hyperthyroidism (p > 0.05). Sensitivity analyses confirmed the robustness of these results, suggesting minimal susceptibility to confounding factors and bias.Conclusion: The finding that longer telomeres reduce hyperthyroidism risk highlights the need to validate hyperthyroidism’s impact on telomere length, offering valuable insights for prevention and treatment.


  • Top Cited Papers
    Open Access

    Qwen-2.5 Outperforms Other Large Language Models in the Chinese National Nursing Licensing Examination: Retrospective Cross-Sectional Comparative Study

    by Shiben Zhu, Wanqin Hu, Zhi Yang, Jiani Yan, Fang Zhang

    Background: Large language models (LLMs) have been proposed as valuable tools in medical education and practice. The Chinese National Nursing Licensing Examination (CNNLE) presents unique challenges for LLMs due to its requirement for both deep domain–specific nursing knowledge and the ability to make complex clinical decisions, which differentiates it from more general medical examinations. However, their potential application in the CNNLE remains unexplored. Objective: This study aims to evaluates the accuracy of 7 LLMs including GPT-3.5, GPT-4.0, GPT-4o, Copilot, ERNIE Bot-3.5, SPARK, and Qwen-2.5 on the CNNLE, focusing on their ability to handle domain-specific nursing knowledge and clinical decision-making. We also explore whether combining their outputs using machine learning techniques can improve their overall accuracy. Methods: This retrospective cross-sectional study analyzed all 1200 multiple-choice questions from the CNNLE conducted between 2019 and 2023. Seven LLMs were evaluated on these multiple-choice questions, and 9 machine learning models, including Logistic Regression, Support Vector Machine, Multilayer Perceptron, k-nearest neighbors, Random Forest, LightGBM, AdaBoost, XGBoost, and CatBoost, were used to optimize overall performance through ensemble techniques. Results: Qwen-2.5 achieved the highest overall accuracy of 88.9%, followed by GPT-4o (80.7%), ERNIE Bot-3.5 (78.1%), GPT-4.0 (70.3%), SPARK (65.0%), and GPT-3.5 (49.5%). Qwen-2.5 demonstrated superior accuracy in the Practical Skills section compared with the Professional Practice section across most years. It also performed well in brief clinical case summaries and questions involving shared clinical scenarios. When the outputs of the 7 LLMs were combined using 9 machine learning models, XGBoost yielded the best performance, increasing accuracy to 90.8%. XGBoost also achieved an area under the curve of 0.961, sensitivity of 0.905, specificity of 0.978, F1-score of 0.901, positive predictive value of 0.901, and negative predictive value of 0.977. Conclusions: This study is the first to evaluate the performance of 7 LLMs on the CNNLE and that the integration of models via machine learning significantly boosted accuracy, reaching 90.8%. These findings demonstrate the transformative potential of LLMs in revolutionizing health care education and call for further research to refine their capabilities and expand their impact on examination preparation and professional training.


  • Open Access

    Casual effects of telomere length on sarcoidosis: a bidirectional Mendelian randomization analysis

    by Shiben Zhu, Ziyu Hao, Qihang Chen, Xiaoliu Liu, Wenyan Wu, Yanping Luo, Fang Zhang

    Background: Telomere length, crucial for genomic stability, have been implicated in various inflamm-aging diseases, but their role in sarcoidosis remains unexplored.Objective: This study aims to explore the casual effects between TL and sarcoidosis via a bidirectional Mendelian Randomization (MR) study.Methods: We examined single nucleotide polymorphisms (SNPs) associated with TL and sarcoidosis, utilizing available open-access genome-wide association study (GWAS) databases from the UK Biobank and FinnGen. We employed five MR techniques, including Inverse Variance Weighted (IVW), MR Egger, weighted median (WM), Robust adjusted profile score (RAPS), and Maximum likelihood, to assess causal relationships and explore pleiotropy.Results: Summary data extracted from GWAS datasets of TL (n = 472,174) and (n = 217,758) of European ancestry. Employing 130 SNPs with genome-wide significance as instrumental factors for TL, we detect a significant negative correlation between TL and sarcoidosis (OR: 0.682, 95% confidence interval: 0.524–0.888, p: 0.0045). Similarly, utilizing 6 SNPs with genome-wide significance as instrumental factors for sarcoidosis, we fail to identify a noteworthy association between sarcoidosis and TL (OR: 0.992, 95% confidence interval: 0.979–1.005, p: 0.2424).Conclusion: Our results suggest that longer telomeres may reduce the risk of sarcoidosis, highlighting TL as a potential biomarker for diagnosis and long-term monitoring. Understanding the critical role of telomere shortening enables more effective focus on diagnosing, treating, and curing sarcoidosis linked to telomeres. Clinical investigations into treatments that enhance TL are warranted.


  • Open Access

    Qwen-2.5 Outperforms Other Large Language Models in the Chinese National Nursing Licensing Examination: Retrospective Cross-Sectional Comparative Study (Preprint)

    by Shiben Zhu, Wanqin Hu, Zhi Yang, Jiani Yan, Fang Zhang

    BACKGROUND Large language models (LLMs) have been proposed as valuable tools in medical education and practice. The Chinese National Nursing Licensing Examination (CNNLE) presents unique challenges for LLMs due to its requirement for both deep domain–specific nursing knowledge and the ability to make complex clinical decisions, which differentiates it from more general medical examinations. However, their potential application in the CNNLE remains unexplored. OBJECTIVE This study aims to evaluates the accuracy of 7 LLMs including GPT-3.5, GPT-4.0, GPT-4o, Copilot, ERNIE Bot-3.5, SPARK, and Qwen-2.5 on the CNNLE, focusing on their ability to handle domain-specific nursing knowledge and clinical decision-making. We also explore whether combining their outputs using machine learning techniques can improve their overall accuracy. METHODS This retrospective cross-sectional study analyzed all 1200 multiple-choice questions from the CNNLE conducted between 2019 and 2023. Seven LLMs were evaluated on these multiple-choice questions, and 9 machine learning models, including Logistic Regression, Support Vector Machine, Multilayer Perceptron, k-nearest neighbors, Random Forest, LightGBM, AdaBoost, XGBoost, and CatBoost, were used to optimize overall performance through ensemble techniques. RESULTS Qwen-2.5 achieved the highest overall accuracy of 88.9%, followed by GPT-4o (80.7%), ERNIE Bot-3.5 (78.1%), GPT-4.0 (70.3%), SPARK (65.0%), and GPT-3.5 (49.5%). Qwen-2.5 demonstrated superior accuracy in the Practical Skills section compared with the Professional Practice section across most years. It also performed well in brief clinical case summaries and questions involving shared clinical scenarios. When the outputs of the 7 LLMs were combined using 9 machine learning models, XGBoost yielded the best performance, increasing accuracy to 90.8%. XGBoost also achieved an area under the curve of 0.961, sensitivity of 0.905, specificity of 0.978, F1-score of 0.901, positive predictive value of 0.901, and negative predictive value of 0.977. CONCLUSIONS This study is the first to evaluate the performance of 7 LLMs on the CNNLE and that the integration of models via machine learning significantly boosted accuracy, reaching 90.8%. These findings demonstrate the transformative potential of LLMs in revolutionizing health care education and call for further research to refine their capabilities and expand their impact on examination preparation and professional training.

  • Open Access

    Evolution of Chatbots in Nursing Education: Narrative Review

    by Fang Zhang, Xiaoliu Liu, Wenyan Wu, Shiben Zhu

    Background: The integration of chatbots in nursing education is a rapidly evolving area with potential transformative impacts. This narrative review aims to synthesize and analyze the existing literature on chatbots in nursing education. Objective: This study aims to comprehensively examine the temporal trends, international distribution, study designs, and implications of chatbots in nursing education. Methods: A comprehensive search was conducted across 3 databases (PubMed, Web of Science, and Embase) following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram. Results: A total of 40 articles met the eligibility criteria, with a notable increase of publications in 2023 (n=28, 70%). Temporal analysis revealed a notable surge in publications from 2021 to 2023, emphasizing the growing scholarly interest. Geographically, Taiwan province made substantial contributions (n=8, 20%), followed by the United States (n=6, 15%) and South Korea (n=4, 10%). Study designs varied, with reviews (n=8, 20%) and editorials (n=7, 18%) being predominant, showcasing the richness of research in this domain. Conclusions: Integrating chatbots into nursing education presents a promising yet relatively unexplored avenue. This review highlights the urgent need for original research, emphasizing the importance of ethical considerations.


  • Open Access

    Casual Effects of Telomere Length on Sarcoidosis: A Bidirectional Mendelian Randomisation Analysis

    by Shiben Zhu, Ziyu Hao, Qihang Chen, Xiaoliu Liu, Wenyan Wu, Yanping Luo, Fang Zhang

    Background Telomere length (TL), crucial for genomic stability, have been implicated in various inflamm-aging diseases, but their role in sarcoidosis remains unexplored. Objective This study aims to explore the casual effects between TL and sarcoidosis via a bidirectional Mendelian Randomization (MR) study. Methods We examined single nucleotide polymorphisms (SNPs) associated with TL and sarcoidosis, utilizing available open-access genome-wide association study (GWAS) databases from the UK Biobank and FinnGen. We employed five MR techniques, including Inverse Variance Weighted (IVW), MR Egger, weighted median (WM), Robust adjusted profile score (RAPS), and Maximum likelihood, to assess causal relationships and explore pleiotropy. Results Summary data extracted from GWAS datasets of TL (n = 472,174) and (n = 217,758) of European ancestry. Employing 130 SNPs with genome-wide significance as instrumental factors for TL, we detect a significant negative correlation between TL and sarcoidosis (OR: 0.682, 95% confidence interval: 0.524–0.888,p:0.0045). Similarly, utilizing 6 SNPs with genome-wide significance as instrumental factors for sarcoidosis, we fail to identify a noteworthy association between sarcoidosis and TL (OR:0.992, 95% confidence interval: 0.979–1.005, p: 0.2424). Conclusion Our results suggest that longer telomeres may reduce the risk of sarcoidosis, highlighting TL as a potential biomarker for diagnosis and long-term monitor. This study contributes to understanding the genetic interplay between TL and sarcoidosis and opens new avenues for prevention and therapeutic interventions.