XKXiaomin Kang

Xiaomin Kang

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

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Xiaomin Kang has a total of 30 co-authors. Here are the top co-authors by number of shared publications:

  • Fengmei Lian

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China

    7 shared publications
  • Linlin Jiang

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, China

    5 shared publications
  • Rongrong Zhou

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, People’s Republic of China

    4 shared publications
  • Xuedong An

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, People’s Republic of China

    4 shared publications
  • Yingying Duan

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, People’s Republic of China

    4 shared publications
  • Yuting Sun

    Guang’anmen Hospital, China Academy of Chinese Medical Sciences, Beijing, People’s Republic of China

    4 shared publications

Latest Publications

  • Open Access

    Quality and readability of large language model–generated responses on type 2 diabetes remission: a cross-sectional balanced factorial study

    by Wenjun Wu, Nana Qin, Xiaomin Kang, Linlin Jiang, Qing Gao, Chenran Liu, Zehua Li, Shiyu Liao, Guannan Li, Jinglun Li, et al.

    Large language models (LLMs) are increasingly being explored for patient education and medical information delivery. However, their outputs may involve trade-offs among medical accuracy, comprehensiveness, empathy, and readability. Type 2 diabetes remission is a clinically nuanced concept that is readily misunderstood by patients, making it a suitable context in which to evaluate both the medical quality and patient accessibility of artificial intelligence–generated health information. To systematically compare the medical quality and readability of responses concerning type 2 diabetes remission generated by different LLMs and prompting strategies, and to characterize their joint quality–readability profiles. We conducted a cross-sectional study using a balanced factorial design. Responses were generated for 16 core questions concerning type 2 diabetes remission using 4 LLMs—DeepSeek-V3.2,GPT-5.5,Gemini 3.1 Pro and Claude Sonnet 4.6—and 6 prompting strategies: Keyword, Question, 6th-grade, Guideline, Patient, and Doctor. This yielded 384 AI-generated responses. Two endocrinologists, each with more than 5 years of clinical experience, independently evaluated medical quality using a prespecified rubric. Five domains—Accuracy, Relevance, Comprehensiveness, Empathy, and Falsehood avoidance—were rated on 5-point ordinal scales. A descriptive Composite quality score was also calculated. Readability was assessed using the Flesch Reading Ease Score (FRES), Flesch-Kincaid Grade Level (FKGL), Gunning Fog Index (GFOG), and Simple Measure of Gobbledygook (SMOG). Quality ratings were analyzed using cumulative link mixed models, whereas readability outcomes were analyzed using linear mixed-effects models; Question ID was included as a random intercept in all models. Exploratory analyses examined Spearman correlations between readability and quality and identified Pareto-efficient model–prompt combinations. All 384 responses were included in the analysis, with no missing data. The descriptive Composite quality score ranged from 4.175 for Gemini 3.1 Pro to 4.285 for DeepSeek-V3.2 across models and from 4.066 for the 6th-grade strategy to 4.350 for the Doctor strategy across prompting strategies. Cumulative link mixed models showed significant main effects of model across all 5 quality domains: Accuracy (LR χ² = 10.964, LR-based pseudo-R² = 0.0281, 95% CI 0.0015–0.0636, P = 0.012), Relevance (LR χ² = 54.042, LR-based pseudo-R² = 0.1313, 95% CI 0.0688–0.1974, P < 0.001), Comprehensiveness (LR χ² = 12.018, LR-based pseudo-R² = 0.0308, 95% CI 0.0025–0.0677, P = 0.007), Empathy (LR χ² = 50.905, LR-based pseudo-R² = 0.1242, 95% CI 0.0632–0.1890, P < 0.001), and Falsehood avoidance (LR χ² = 18.752, LR-based pseudo-R² = 0.0477, 95% CI 0.0107–0.0922, P < 0.001). Prompting strategy also showed significant main effects across the quality domains, with the largest effect observed for Empathy (LR χ² = 243.836, LR-based pseudo-R² = 0.4701, 95% CI 0.3786–0.5477, P < 0.001). Descriptively, GPT-5.5 had the numerically highest mean scores for Accuracy (4.448) and Falsehood avoidance (4.469), whereas DeepSeek-V3.2 had the numerically highest mean scores for Relevance (4.948), Comprehensiveness (4.427), and Empathy (3.615). Among prompting strategies, Patient had the numerically highest mean Empathy score (4.359), whereas Doctor had the numerically highest mean Comprehensiveness score (4.609). For readability, both model and prompting strategy showed significant main effects on FRES, FKGL, GFOG, and SMOG. DeepSeek-V3.2 had the numerically highest mean FRES (51.28), whereas Claude Sonnet 4.6 had the lowest (32.98); among prompting strategies, 6th-grade had the numerically highest mean FRES (69.70), whereas Doctor had the lowest (27.50). In an exploratory response-level correlation analysis that did not account for clustering within Question ID and was intended for hypothesis generation only, FRES was positively correlated with Empathy (ρ = 0.53) and negatively correlated with Comprehensiveness, Falsehood avoidance, and the Composite quality score (ρ = −0.46, − 0.40, and − 0.26, respectively; FDR-adjusted q < 0.001), indicating directionally divergent association patterns between readability and different dimensions of medical quality. In patient-education responses related to type 2 diabetes remission, LLM output performance was influenced by both model choice and prompting strategy. No single model–prompt combination consistently outperformed the others across all evaluated outcomes. Within the T2DM remission question set examined in this study, response optimization required trade-offs among accuracy, comprehensiveness, readability, falsehood avoidance, and empathy.

  • Open Access

    Programmed Cell Death in Diabetic Kidney Disease: Mechanisms and Therapeutic Targeting

    by Shanshan Tang, Yuting Sun, Wenjie Sun, Xiaomin Kang, Xuefei Zhao, Linlin Jiang, Qing Gao, Xuedong An, Hangyu Ji, Fengmei Lian

    The escalating incidence and mortality of diabetic kidney disease (DKD) underscore the critical need to elucidate its pathogenesis. Programmed cell death (PCD) plays a dual role in maintaining physiological homeostasis and driving pathological processes in DKD. Accumulating evidence demonstrates that apoptosis, autophagy, pyroptosis, and ferroptosis contribute directly or indirectly to DKD progression via distinct gene-regulated signaling pathways. Recently identified PCD modes (eg, necroptosis, parthanatos) remain poorly characterized in DKD, with emerging evidence suggesting crosstalk between different PCD pathways. This review synthesizes current knowledge on PCD-mediated DKD pathogenesis and PCD-targeted therapies, while highlighting research limitations (eg, unclear PCD interactions, translational gaps). We propose that dissecting the multifaceted roles of PCD in DKD will deepen mechanistic understanding and accelerate the development of novel therapeutics, offering significant scientific and clinical benefits.


  • Open Access

    Response to Professor Hu Jinyu’s Letter [Response to Letter]

    by Xiaomin Kang, De Jin, Hangyu Ji, Xuedong An, Yuehong Zhang, Liyun Duan, Cunqing Yang, Rongrong Zhou, Yingying Duan, Yuqing Zhang, et al.

    Response to Letter in regards to Comment on the Use of Gegen Qinlian Decoction to Treat Type 2 Diabetes Mellitus [Letter]

  • Open Access

    The Clinical Efficacy of Gegen Qinlian Decoction in Treating Type 2 Diabetes is Positively Correlated with the Dose of Coptidis rhizoma: Three Randomized, Doubleblind, Dose-Parallel Controlled Clinical Trials

    by Xiaomin Kang, De Jin, Hangyu Ji, Xuedong An, Yuehong Zhang, Liyun Duan, Cunqing Yang, Rongrong Zhou, Yingying Duan, Yuqing Zhang, et al.

    Background: The results of Study 1 we have published proved that medium and high doses of Gegen Qinlian Decoction (GQD) were effective in treating type 2 diabetes (T2D) with damp-heat syndrome. However, whether the main drug of GQD in treating T2D was Puerariae Lobatae Radix or Coptidis Rhizoma has always been a hot topic of debate among many doctors. Therefore, we conducted Study 2 and Study 3 to determine the main drug of GQD for treating T2D. Methods: Both Study 2 and Study 3 were randomized, double-blind, dose-parallel controlled, multicenter trials. In Study 2, Puerariae Lobatae Radix was used as the main drug, and in Study 3, Coptidis Rhizoma was used as the main drug. About 120 patients with newly diagnosed T2D were enrolled in each study and randomized 1:1:1 to three treatment groups. The three treatment groups were named HD, MD, and LD groups according to the high, medium, and low doses of the main drug. The course of treatment was 12 weeks. The primary outcomes were the changes in HbA1c. Results: In Study 2, the HbA1c decreased by 0.58 (0.87), 0.28 (1.17), and 0.55 (0.85) in the HD, MD, and LD groups, respectively, with no significant difference between treatment groups according to covariance analysis (F=0.66, P=0.5206). In Study 3, the HbA1c decreased by 0.75 (0.82), 0.34 (0.71), and 0.26 (0.79) in the HD, MD, and LD groups respectively. By analysis of covariance, the change values of HbA1c were significantly different among the three groups (F=3.11, P=0.0492). Conclusion: The changes in HbA1c were positively correlated with the dose of Coptidis Rhizoma, but not significantly with the dose of Puerariae Lobatae Radix. It demonstrated that the main drug of GQD in treating T2D patients is Coptidis Rhizoma.


  • Top Cited Papers
    Open Access

    Efficacy and mechanisms of traditional Chinese medicine for COVID-19: a systematic review

    by Xiaomin Kang, De Jin, Linlin Jiang, Yuqing Zhang, Yuehong Zhang, Xuedong An, Liyun Duan, Cunqing Yang, Rongrong Zhou, Yingying Duan, et al.

    Since the outbreak of coronavirus disease 2019 (COVID-19), traditional Chinese medicine (TCM) has made an important contribution to the prevention and control of the epidemic. This review aimed to evaluate the efficacy and explore the mechanisms of TCM for COVID-19. We systematically searched 7 databases from their inception up to July 21, 2021, to distinguish randomized controlled trials (RCTs), cohort studies (CSs), and case–control studies (CCSs) of TCM for COVID-19. Two reviewers independently completed the screening of literature, extraction of data, and quality assessment of included studies. Meta-analysis was performed using Review Manager 5.4 software. Eventually, 29 RCTs involving 3060 patients and 28 retrospective studies (RSs) involving 12,460 patients were included. The meta-analysis demonstrated that TCM could decrease the proportion of patients progressing to severe cases by 55% and the mortality rate of severe or critical patients by 49%. Moreover, TCM could relieve clinical symptoms, curtail the length of hospital stay, improve laboratory indicators, and so on. In addition, we consulted the literature and obtained 149 components of Chinese medicinal herbs that could stably bind to antiviral targets or anti-inflammatory or immune-regulating targets by the prediction of molecular docking. It suggested that the mechanisms involved anti-virus, anti-inflammation, and regulation of immunity. Our study made a systematic review on the efficacy of TCM for COVID-19 and discussed the possible mechanisms, which provided clinical reference and theoretical basis for further research on the mechanism of TCM for COVID-19.


  • Open Access

    Investigating the Mechanisms of Pollen Typhae in the Treatment of Diabetic Retinopathy Based on Network Pharmacology and Molecular Docking

    by Rongrong Zhou, De Jin, Yuqing Zhang, Liyun Duan, Yuehong Zhang, Yingying Duan, Xiaomin Kang, Fengmei Lian

    Objective. To explore the main bioactive compounds and investigate the underlying mechanism of Pollen Typhae (PT) against diabetic retinopathy (DR) by network pharmacology and molecular docking analysis. Methods. Bioactive ingredients and the target proteins of PT were obtained from TCMSP, and the related target genes were acquired from the SwissTargetPrediction database. The target genes of DR were obtained from GeneCards, TTD database, DisGeNET database, and DrugBank. The compound-target interaction network was established based on Cytoscape 3.7.2. The protein-protein interaction (PPI) network was constructed via STRING database and Cytoscape 3.7.2. Gene ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis were visualized through DAVID database and Bioinformatics. Ingredient-gene-pathway network analysis was conducted to further screen the ingredients, target proteins, and pathways closely related to the biological mechanism on PT for DR, and molecular docking analysis was performed by SYBYL-X 2.1.1 software. Finally, the mechanism and underlying targets of PT in the treatment of DR were predicted. Results. A total of 8 compounds and 171 intersection targets were obtained based on the online network database. 7 main compounds were screened from compound-target network, and 53 targets including the top six key targets (PTGS2, AKT1, VEGFA, MAPK3, TNF, and EGFR) were further acquired from PPI analysis. The 53 key targets covered 80 signaling pathways, among which PI3K-Akt signaling pathway, focal adhesion, Rap1 signaling pathway, VEGF signaling pathway, and HIF-1 signaling pathway were closely connected with the biological mechanism involved in the alleviation of DR by PT. Ingredient-gene-pathway network shows that AKTI, EGFR, and VEGFA were core genes, kaempferol and isorhamnetin were pivotal ingredients, and VEGF signaling pathway and Rap1 signaling pathway were closely involved in anti-DR. The docking results indicated that five main compounds (arachidonic acid, isorhamnetin, quercetin, kaempferol, and (2R)-5,7-dihydroxy-2-(4-hydroxyphenyl)chroman-4-one) had good binding activity with EGFR and AKT1 targets. Conclusion. The active ingredients in PT may regulate the levels of inflammatory factors, suppress the oxidative stress, and inhibit the proliferation, migration, and invasion of retinal pericytes by acting on PTGS2, AKT1, VEGFA, MAPK3, TNF, and EGFR targets through VEGF signaling pathway, PI3K-Akt signaling pathway, Rap1 signaling pathway, and HIF-1 signaling pathway to play a therapeutic role in diabetic retinopathy.


  • Open Access

    Systematic Elucidation of the Mechanism of Sappan Lignum in the Treatment of Diabetic Peripheral Neuropathy Based on Network Pharmacology

    by Xiaomin Kang, De Jin, Yuqing Zhang, Rongrong Zhou, Yuehong Zhang, Fengmei Lian

    Background. Diabetic peripheral neuropathy (DPN) is one of the most common chronic complications of diabetes, which seriously affects the physical and mental health of patients. Sappan Lignum (SL) is effective in treating DPN. Previous reports have shown that SL has a clear hypoglycemic and anti-inflammatory effect. However, the study of SL in the treatment of DPN is still limited and rare. Objective. To investigate the mechanism of SL in the treatment of DPN based on network pharmacology. Methods. The active ingredients of SL were screened by related databases. The compound targets were collected by the target prediction platforms. The DPN-related targets were gathered through disease databases. The intersection targets were obtained by uploading the compound targets and disease targets to Venny 2.1.0, and a compound-target network was constructed by Cytoscape3.7.2. The protein-protein interaction (PPI) relationships were obtained by the STRING11.0 database. Genome Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed using the DAVID6.8 database. Molecular docking of key compounds and core targets was performed by DockThor. Results. A total of 29 compounds and 51 intersection targets with potential therapeutic effects on DPN were obtained. The compound-target network construction resulted in four key compounds: protostemonine, 3-deoxysappanchalcone, 7,3′,4′-trihydroxy-3-benzyl-2H-chromene, and o-12′-methylergocornine. PPI network analysis yielded 10 core targets: AKT1, MAPK3, CXCL8, TNF, OPRM1, MTOR, STAT3, MAPK8, SIRT1, and HSP90AA1. KEGG analysis resulted in 82 signaling pathways ( P < 0.05 ), including insulin resistance, HIF-1 signaling pathway, and type II diabetes. The docking results indicated that the main active compounds could stably bind to core targets. Conclusion. SL had the mechanism of multiple ingredients, multiple targets, and multiple pathways in the treatment of DPN. This study provided a scientific basis for further research on the treatment of DPN with SL and its extracts.