Xu Tong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xu Tong has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
Jiatian Yu
Institute of Basic Theory for Chinese Medicine, China Academy of Chinese Medical Sciences, Beijing, China
2 shared publications- Chao Sun1 shared publication
Conghui Zhou
Affiliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, China
1 shared publication- Conghui Zhou
Affiliated Hospital of Integrated Traditional Chinese and Western Medicine, Nanjing University of Chinese Medicine, Nanjing, China
1 shared publication Gaoshuang Fu
1 shared publication- Jack H. Culbert
GESIS – Leibniz Institute for the Social Sciences
1 shared publication
Latest Publications
- Open Access
- 27 July 2026
- Published by Wolters Kluwer Health in Guidelines and Standards of Chinese Medicine
- Vol. 4 (3) , 207-212
Leveraging multibiomarkers for precision diagnosis of coronary heart disease with phlegm and blood stasis syndrome: a real-world observational study protocol
Objective: To assess the diagnostic value of multiple biomarkers, including blood lipid, coagulation, and inflammatory markers, both individually and in combination, for assisting the precision diagnosis of coronary heart disease with phlegm and blood stasis syndrome (CHD-PBSS) in a real-world observational study, thereby providing a scientific basis for future research on the precision diagnosis of CHD-PBSS. Methods: This is a multicenter, real-world observational study designed to recruit 360 patients aged 35 to 70 years with chronic stable CHD from several medical institutions, including Beijing First Hospital of Integrated Chinese and Western Medicine, Yichang Hospital of Traditional Chinese Medicine (TCM), and Huanggang Hospital of TCM. Patients with CHD-PBSS were assigned to the PBSS group ( n = 180), whereas those with CHD but without PBSS (CHD-NPBSS) were assigned to the NPBSS group (n=180). We compared the levels of blood lipid, coagulation, and inflammatory biomarkers between the 2 groups and subsequently evaluated the diagnostic accuracy of these biomarkers, both individually and in combination, for identifying PBSS in patients with CHD. Discussion: The precision diagnosis of PBSS, a critical syndrome in CHD, is crucial for developing effective prevention and management strategies within the framework of TCM. This study aims to identify clinical diagnostic biomarkers of CHD-PBSS through real-world clinical trials, providing a theoretical basis for achieving accurate TCM syndrome diagnosis. However, the observational design has inherent limitations, as it only offers basic demographic information and does not allow for causal inference. In addition, the complexity of TCM syndromes and the subjectivity of clinicians’ judgements may introduce information and selection biases, potentially affecting the accuracy of the results.
- 27 July 2026
- Published by Wolters Kluwer Health in Guidelines and Standards of Chinese Medicine
- Vol. 4 (3) , 207-212
Open AccessLeveraging multibiomarkers for precision diagnosis of coronary heart disease with phlegm and blood stasis syndrome: a real-world observational study protocol
Objective: To assess the diagnostic value of multiple biomarkers, including blood lipid, coagulation, and inflammatory markers, both individually and in combination, for assisting the precision diagnosis of coronary heart disease with phlegm and blood stasis syndrome (CHD-PBSS) in a real-world observational study, thereby providing a scientific basis for future research on the precision diagnosis of CHD-PBSS. Methods: This is a multicenter, real-world observational study designed to recruit 360 patients aged 35 to 70 years with chronic stable CHD from several medical institutions, including Beijing First Hospital of Integrated Chinese and Western Medicine, Yichang Hospital of Traditional Chinese Medicine (TCM), and Huanggang Hospital of TCM. Patients with CHD-PBSS were assigned to the PBSS group ( n = 180), whereas those with CHD but without PBSS (CHD-NPBSS) were assigned to the NPBSS group (n=180). We compared the levels of blood lipid, coagulation, and inflammatory biomarkers between the 2 groups and subsequently evaluated the diagnostic accuracy of these biomarkers, both individually and in combination, for identifying PBSS in patients with CHD. Discussion: The precision diagnosis of PBSS, a critical syndrome in CHD, is crucial for developing effective prevention and management strategies within the framework of TCM. This study aims to identify clinical diagnostic biomarkers of CHD-PBSS through real-world clinical trials, providing a theoretical basis for achieving accurate TCM syndrome diagnosis. However, the observational design has inherent limitations, as it only offers basic demographic information and does not allow for causal inference. In addition, the complexity of TCM syndromes and the subjectivity of clinicians’ judgements may introduce information and selection biases, potentially affecting the accuracy of the results.
- Open Access
- 7 October 2024
- Published by Wolters Kluwer Health in Guidelines and Standards of Chinese Medicine
- Vol. 2 (4) , 174-181
Research progress on knowledge discovery in traditional Chinese medicine medical records based on artificial intelligence
Traditional Chinese Medicine (TCM) medical records represent the clinical practice records of physicians, encapsulating a wealth of TCM clinical diagnosis and treatment experiences as well as academic thoughts of numerous TCM practitioners. Such various information provides certain basic conditions for the integration of artificial intelligence (AI) technology with TCM. Through analyzing successful clinical records, AI technology can extract practical and understandable knowledge. Therefore, this paper summarizes and analyzes the application of information technology in the field of knowledge discovery research from Chinese medical records. It focuses on 3 key aspects: the resources of TCM Medical Records, the objects of TCM records mining, and the application of AI technology. The aim is to provide beneficial references for maximizing the value of TCM medical records.
- 7 October 2024
- Published by Wolters Kluwer Health in Guidelines and Standards of Chinese Medicine
- Vol. 2 (4) , 174-181
Open AccessResearch progress on knowledge discovery in traditional Chinese medicine medical records based on artificial intelligence
Traditional Chinese Medicine (TCM) medical records represent the clinical practice records of physicians, encapsulating a wealth of TCM clinical diagnosis and treatment experiences as well as academic thoughts of numerous TCM practitioners. Such various information provides certain basic conditions for the integration of artificial intelligence (AI) technology with TCM. Through analyzing successful clinical records, AI technology can extract practical and understandable knowledge. Therefore, this paper summarizes and analyzes the application of information technology in the field of knowledge discovery research from Chinese medical records. It focuses on 3 key aspects: the resources of TCM Medical Records, the objects of TCM records mining, and the application of AI technology. The aim is to provide beneficial references for maximizing the value of TCM medical records.
- Open Access
- 7 November 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Utilizing Large Language Models for Named Entity Recognition in Traditional Chinese Medicine against COVID-19 Literature: Comparative Study (Preprint)
BACKGROUND Recent advances in large language models (LLMs) have shown remarkable performance on various downstream tasks in zero- and few-shot scenarios, shedding light on named entity recognition (NER) in low-resource domains. Traditional Chinese medicine (TCM) against COVID-19 has been a new research topic and led to niche research literature. NER techniques are crucial for extracting and utilizing the rich knowledge in such literature. OBJECTIVE To explore and compare the performance of ChatGPT and other state-of-the-art LLMs on domain-specific NER tasks covering different entity types and domains in TCM against COVID-19 literature. METHODS We established a dataset of 389 articles on TCM against COVID-19, and manually annotated 48 of them with 6 types of entities belonging to 3 domains as the ground truth, against which the NER performance of LLMs can be assessed. We then performed NER tasks for the 6 entity types using ChatGPT (GPT-3.5 and GPT-4) and 4 state-of-the-art BERT-based question-answering (QA) models (RoBERTa, MiniLM, PubMedBERT and SciBERT) without prior training on the specific task. A domain fine-tuned model (GSAP-NER) was also applied for a comprehensive comparison on one of the entity types. In the task setup, we considered two different matching methods, namely exact match and fuzzy match, to have better coverage of potential application scenarios. The evaluation metrics we used are Precision, Recall, and F-1 in both exact match and fuzzy match. RESULTS The overall performance of LLMs varied significantly in exact match and fuzzy match. In the fuzzy match, ChatGPT surpassed BERT-based QA models in 5 out of 6 tasks, while in exact match, BERT-based QA models outperformed ChatGPT in 5 out of 6 tasks but with a smaller F-1 difference. GPT-4 showed a significant advantage over other models in fuzzy match, especially on the entity type of TCM formula and the Chinese patent drug (TFD) and ingredient (IG), achieving a higher F-1 score of 0.814 and 0.689, respectively. Although GPT-4 outperformed BERT-based models on entity type of herb (HB, 0.324), target (TG, 0.319), and research method (RM, 0.443), none of the F-1 scores exceeded 0.5. GSAP-NER, outperformed GPT-4 in terms of F-1 by a slight margin (0.45 vs 0.433) on RM. ChatGPT achieved considerably higher recalls than precisions, particularly in the fuzzy match. CONCLUSIONS The NER performance of LLMs is highly dependent on the entity type, and their performance varies across application scenarios. ChatGPT could be a good choice for scenarios where high recall is favored, e.g., for domain novices obtaining an extensive overview of the field. However, for knowledge acquisition in rigorous scenarios, neither ChatGPT nor BERT-based QA models are off-the-shelf tools for professional practitioners. Besides, these BERT-based models are open-source, and thus more accessible for scientific inquiry and hence worth further exploration.
- 7 November 2023
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessUtilizing Large Language Models for Named Entity Recognition in Traditional Chinese Medicine against COVID-19 Literature: Comparative Study (Preprint)
BACKGROUND Recent advances in large language models (LLMs) have shown remarkable performance on various downstream tasks in zero- and few-shot scenarios, shedding light on named entity recognition (NER) in low-resource domains. Traditional Chinese medicine (TCM) against COVID-19 has been a new research topic and led to niche research literature. NER techniques are crucial for extracting and utilizing the rich knowledge in such literature. OBJECTIVE To explore and compare the performance of ChatGPT and other state-of-the-art LLMs on domain-specific NER tasks covering different entity types and domains in TCM against COVID-19 literature. METHODS We established a dataset of 389 articles on TCM against COVID-19, and manually annotated 48 of them with 6 types of entities belonging to 3 domains as the ground truth, against which the NER performance of LLMs can be assessed. We then performed NER tasks for the 6 entity types using ChatGPT (GPT-3.5 and GPT-4) and 4 state-of-the-art BERT-based question-answering (QA) models (RoBERTa, MiniLM, PubMedBERT and SciBERT) without prior training on the specific task. A domain fine-tuned model (GSAP-NER) was also applied for a comprehensive comparison on one of the entity types. In the task setup, we considered two different matching methods, namely exact match and fuzzy match, to have better coverage of potential application scenarios. The evaluation metrics we used are Precision, Recall, and F-1 in both exact match and fuzzy match. RESULTS The overall performance of LLMs varied significantly in exact match and fuzzy match. In the fuzzy match, ChatGPT surpassed BERT-based QA models in 5 out of 6 tasks, while in exact match, BERT-based QA models outperformed ChatGPT in 5 out of 6 tasks but with a smaller F-1 difference. GPT-4 showed a significant advantage over other models in fuzzy match, especially on the entity type of TCM formula and the Chinese patent drug (TFD) and ingredient (IG), achieving a higher F-1 score of 0.814 and 0.689, respectively. Although GPT-4 outperformed BERT-based models on entity type of herb (HB, 0.324), target (TG, 0.319), and research method (RM, 0.443), none of the F-1 scores exceeded 0.5. GSAP-NER, outperformed GPT-4 in terms of F-1 by a slight margin (0.45 vs 0.433) on RM. ChatGPT achieved considerably higher recalls than precisions, particularly in the fuzzy match. CONCLUSIONS The NER performance of LLMs is highly dependent on the entity type, and their performance varies across application scenarios. ChatGPT could be a good choice for scenarios where high recall is favored, e.g., for domain novices obtaining an extensive overview of the field. However, for knowledge acquisition in rigorous scenarios, neither ChatGPT nor BERT-based QA models are off-the-shelf tools for professional practitioners. Besides, these BERT-based models are open-source, and thus more accessible for scientific inquiry and hence worth further exploration.
- Open Access
- 1 April 2022
- Published by AME Publishing Company in Journal of Thoracic Disease
- Vol. 14 (4) , 1232-1242
The efficacy and safety of Wulingsan modified formulas for chronic heart failure patients: a systematic review and meta-analysis
Background: Chronic heart failure (CHF) is one of the most common cardiovascular diseases, which has caused huge economic burden worldwide. Wulingsan modified formulas have been historically used for CHF in China. However, the efficacy of its treatment for CHF has not been summarized by scholars and lacks of clinical evidence. This study aimed to assess the efficacy and safety of Wulingsan modified formulas for patients with CHF. Methods: A comprehensive literature search was performed in the PubMed, EMBASE, Cochrane Library, Web of Science, Medline (Ovid), China National Knowledge Infrastructure, WanFang, China Science and Technology Journal Database, and SinoMed databases from the date of their inception up to 1st November, 2021. Only randomized controlled trials evaluating Wulingsan modified formulas in patients with CHF were included. The primary outcome of this study was efficacy of Wulingsan modified formulas in the treatment of CHF, and the secondary outcomes included brain natriuretic peptide, left ventricular ejection fractions, and any other changes in the patients’ condition. The risk ratio was applied to evaluate efficiency, and the weighted mean difference (WMD) and 95% confidence interval (CI) were used to merge the continuous variables. The I2 statistic was used to assess the heterogeneity. Sensitivity analysis was used to evaluate whether the single research affected the whole results. Data were extracted by two independent investigators. The Cochrane Risk of Bias tool (version 2.0) was utilized to evaluate the included studies, STATA (version 15.0) was applied for sensitivity analysis, and RevMan 5.3 software was used to conduct the systematic review and meta-analysis. Results: Nineteen studies with a total of 1,631 were included in this meta-analysis. The meta-analysis results were as follows: efficiency, the risk ratio =1.21, 95% CI: 1.15, 1.27; brain natriuretic peptide, WMD =−269.14, 95% CI: −349.25, −189.04; and left ventricular ejection fractions, WMD =8.80, 95% CI: 5.93, 11.68. All of these findings were statistically significant. No statistically significant adverse events were reported in the included articles. Discussion: Wulingsan modified formulas are a reasonable and relatively safe adjuvant therapy for the treatment of CHF.
- 1 April 2022
- Published by AME Publishing Company in Journal of Thoracic Disease
- Vol. 14 (4) , 1232-1242
Open AccessThe efficacy and safety of Wulingsan modified formulas for chronic heart failure patients: a systematic review and meta-analysis
Background: Chronic heart failure (CHF) is one of the most common cardiovascular diseases, which has caused huge economic burden worldwide. Wulingsan modified formulas have been historically used for CHF in China. However, the efficacy of its treatment for CHF has not been summarized by scholars and lacks of clinical evidence. This study aimed to assess the efficacy and safety of Wulingsan modified formulas for patients with CHF. Methods: A comprehensive literature search was performed in the PubMed, EMBASE, Cochrane Library, Web of Science, Medline (Ovid), China National Knowledge Infrastructure, WanFang, China Science and Technology Journal Database, and SinoMed databases from the date of their inception up to 1st November, 2021. Only randomized controlled trials evaluating Wulingsan modified formulas in patients with CHF were included. The primary outcome of this study was efficacy of Wulingsan modified formulas in the treatment of CHF, and the secondary outcomes included brain natriuretic peptide, left ventricular ejection fractions, and any other changes in the patients’ condition. The risk ratio was applied to evaluate efficiency, and the weighted mean difference (WMD) and 95% confidence interval (CI) were used to merge the continuous variables. The I2 statistic was used to assess the heterogeneity. Sensitivity analysis was used to evaluate whether the single research affected the whole results. Data were extracted by two independent investigators. The Cochrane Risk of Bias tool (version 2.0) was utilized to evaluate the included studies, STATA (version 15.0) was applied for sensitivity analysis, and RevMan 5.3 software was used to conduct the systematic review and meta-analysis. Results: Nineteen studies with a total of 1,631 were included in this meta-analysis. The meta-analysis results were as follows: efficiency, the risk ratio =1.21, 95% CI: 1.15, 1.27; brain natriuretic peptide, WMD =−269.14, 95% CI: −349.25, −189.04; and left ventricular ejection fractions, WMD =8.80, 95% CI: 5.93, 11.68. All of these findings were statistically significant. No statistically significant adverse events were reported in the included articles. Discussion: Wulingsan modified formulas are a reasonable and relatively safe adjuvant therapy for the treatment of CHF.