Shurong Gong
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Shurong Gong has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Boyang Zhuang
Fujian Center For Drug Evaluation and Monitoring Fuzhou China
2 shared publications - Cunze Wang
School of Pharmacy Fujian Medical University Fuzhou China
2 shared publications - Jinglin Li
School of Pharmacy Fujian Medical University Fuzhou China
2 shared publications - Ling Wang
Fuzhou University Affiliated Provincial Hospital, Fujian Provincial Hospital, Shengli Clinical Medical College of Fujian Medical University, Fuzhou University, Fuzhou, China
2 shared publications - Meilin Fang
School of Pharmacy Fujian Medical University Fuzhou China
2 shared publications - Mohan Zhang
School of Pharmacy Fujian Medical University Fuzhou China
2 shared publications
Latest Publications
- Top Cited PapersOpen Access
- 30 April 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e64486
Accuracy of Large Language Models When Answering Clinical Research Questions: Systematic Review and Network Meta-Analysis
Background: Large language models (LLMs) have flourished and gradually become an important research and application direction in the medical field. However, due to the high degree of specialization, complexity, and specificity of medicine, which results in extremely high accuracy requirements, controversy remains about whether LLMs can be used in the medical field. More studies have evaluated the performance of various types of LLMs in medicine, but the conclusions are inconsistent. Objective: This study uses a network meta-analysis (NMA) to assess the accuracy of LLMs when answering clinical research questions to provide high-level evidence-based evidence for its future development and application in the medical field. Methods: In this systematic review and NMA, we searched PubMed, Embase, Web of Science, and Scopus from inception until October 14, 2024. Studies on the accuracy of LLMs when answering clinical research questions were included and screened by reading published reports. The systematic review and NMA were conducted to compare the accuracy of different LLMs when answering clinical research questions, including objective questions, open-ended questions, top 1 diagnosis, top 3 diagnosis, top 5 diagnosis, and triage and classification. The NMA was performed using Bayesian frequency theory methods. Indirect intercomparisons between programs were performed using a grading scale. A larger surface under the cumulative ranking curve (SUCRA) value indicates a higher ranking of the corresponding LLM accuracy. Results: The systematic review and NMA examined 168 articles encompassing 35,896 questions and 3063 clinical cases. Of the 168 studies, 40 (23.8%) were considered to have a low risk of bias, 128 (76.2%) had a moderate risk, and none were rated as having a high risk. ChatGPT-4o (SUCRA=0.9207) demonstrated strong performance in terms of accuracy for objective questions, followed by Aeyeconsult (SUCRA=0.9187) and ChatGPT-4 (SUCRA=0.8087). ChatGPT-4 (SUCRA=0.8708) excelled at answering open-ended questions. In terms of accuracy for top 1 diagnosis and top 3 diagnosis of clinical cases, human experts (SUCRA=0.9001 and SUCRA=0.7126, respectively) ranked the highest, while Claude 3 Opus (SUCRA=0.9672) performed well at the top 5 diagnosis. Gemini (SUCRA=0.9649) had the highest rated SUCRA value for accuracy in the area of triage and classification. Conclusions: Our study indicates that ChatGPT-4o has an advantage when answering objective questions. For open-ended questions, ChatGPT-4 may be more credible. Humans are more accurate at the top 1 diagnosis and top 3 diagnosis. Claude 3 Opus performs better at the top 5 diagnosis, while for triage and classification, Gemini is more advantageous. This analysis offers valuable insights for clinicians and medical practitioners, empowering them to effectively leverage LLMs for improved decision-making in learning, diagnosis, and management of various clinical scenarios. Trial Registration: PROSPERO CRD42024558245; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024558245
- 30 April 2025
- Published by JMIR Publications Inc. in Journal of Medical Internet Research
- Vol. 27 (1) , e64486
Top Cited PapersOpen AccessAccuracy of Large Language Models When Answering Clinical Research Questions: Systematic Review and Network Meta-Analysis
Background: Large language models (LLMs) have flourished and gradually become an important research and application direction in the medical field. However, due to the high degree of specialization, complexity, and specificity of medicine, which results in extremely high accuracy requirements, controversy remains about whether LLMs can be used in the medical field. More studies have evaluated the performance of various types of LLMs in medicine, but the conclusions are inconsistent. Objective: This study uses a network meta-analysis (NMA) to assess the accuracy of LLMs when answering clinical research questions to provide high-level evidence-based evidence for its future development and application in the medical field. Methods: In this systematic review and NMA, we searched PubMed, Embase, Web of Science, and Scopus from inception until October 14, 2024. Studies on the accuracy of LLMs when answering clinical research questions were included and screened by reading published reports. The systematic review and NMA were conducted to compare the accuracy of different LLMs when answering clinical research questions, including objective questions, open-ended questions, top 1 diagnosis, top 3 diagnosis, top 5 diagnosis, and triage and classification. The NMA was performed using Bayesian frequency theory methods. Indirect intercomparisons between programs were performed using a grading scale. A larger surface under the cumulative ranking curve (SUCRA) value indicates a higher ranking of the corresponding LLM accuracy. Results: The systematic review and NMA examined 168 articles encompassing 35,896 questions and 3063 clinical cases. Of the 168 studies, 40 (23.8%) were considered to have a low risk of bias, 128 (76.2%) had a moderate risk, and none were rated as having a high risk. ChatGPT-4o (SUCRA=0.9207) demonstrated strong performance in terms of accuracy for objective questions, followed by Aeyeconsult (SUCRA=0.9187) and ChatGPT-4 (SUCRA=0.8087). ChatGPT-4 (SUCRA=0.8708) excelled at answering open-ended questions. In terms of accuracy for top 1 diagnosis and top 3 diagnosis of clinical cases, human experts (SUCRA=0.9001 and SUCRA=0.7126, respectively) ranked the highest, while Claude 3 Opus (SUCRA=0.9672) performed well at the top 5 diagnosis. Gemini (SUCRA=0.9649) had the highest rated SUCRA value for accuracy in the area of triage and classification. Conclusions: Our study indicates that ChatGPT-4o has an advantage when answering objective questions. For open-ended questions, ChatGPT-4 may be more credible. Humans are more accurate at the top 1 diagnosis and top 3 diagnosis. Claude 3 Opus performs better at the top 5 diagnosis, while for triage and classification, Gemini is more advantageous. This analysis offers valuable insights for clinicians and medical practitioners, empowering them to effectively leverage LLMs for improved decision-making in learning, diagnosis, and management of various clinical scenarios. Trial Registration: PROSPERO CRD42024558245; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024558245
- Open Access
- 6 September 2024
- Published by Wolters Kluwer Health in Medicine
- Vol. 103 (36) , e39433
The relationship between mental health problems and risk of infectious diseases: A Mendelian randomization analysis
The causal effects of mental health problems on the risk of infectious diseases remain vague. Investigating them via observational study is challenging as it presents possible confounding factors. Therefore, the objective of this study was to utilize Mendelian randomization (MR) techniques to evaluate the causal relationship between mental health problems and the risk of infectious diseases. Multivariable MR analyses were performed using genome-wide association data for sleep disorders (N = 216,700), depression (N = 500,199), anxiety (N = 290,361), nervous feelings (N = 450,700), unspecified mental disorder (N = 218,792), pneumonia (N = 486,484), skin and subcutaneous tissue infection (SSTI; N = 218,792), intestinal infectious diseases (IIDs; N = 218,792), urinary tract infection (N = 463,010), and central nervous system (CNS) infections (N = 218,792) among individuals of European ancestry. Independent genetic variants significantly (P < 10−8) associated with each exposure were considered instruments. The primary analysis used an inverse variance-weighted method, followed by a series of sensitivity analyses. Genetically predicted sleep disorders were associated with an increased risk of SSTI (odds ratio [OR], 1.29 [95% confidence interval (CI), 1.05–1.59]; P = .017). Genetically predicted depression was linked with an increased risk of CNS infections (OR, 1.59 [95% CI, 1.00–2.53]; P = .049) and SSTI (1.24 [95% CI, 1.03–1.49]; P = .024). Genetically predicted anxiety was associated with IIDs (OR, 1.19 [95% CI, 1.03–1.37]; P = .017) and SSTI (OR, 1.21 [95% CI, 1.02–1.43]; P = .029). There was no significant causal evidence for genetic prediction of nervous feelings and unspecified mental disorders in IIDs, CNS infections, SSTI, pneumonia, or urinary tract infection. Sensitivity analyses showed that the above causal association estimates were robust. In this MR study, we demonstrated a causal relationship between sleep disorders, depression, anxiety, and the risk of infectious diseases. However, no evidence was found to support causality between nervous feelings, unspecified mental disorders, and the risk of infectious diseases.
- 6 September 2024
- Published by Wolters Kluwer Health in Medicine
- Vol. 103 (36) , e39433
Open AccessThe relationship between mental health problems and risk of infectious diseases: A Mendelian randomization analysis
The causal effects of mental health problems on the risk of infectious diseases remain vague. Investigating them via observational study is challenging as it presents possible confounding factors. Therefore, the objective of this study was to utilize Mendelian randomization (MR) techniques to evaluate the causal relationship between mental health problems and the risk of infectious diseases. Multivariable MR analyses were performed using genome-wide association data for sleep disorders (N = 216,700), depression (N = 500,199), anxiety (N = 290,361), nervous feelings (N = 450,700), unspecified mental disorder (N = 218,792), pneumonia (N = 486,484), skin and subcutaneous tissue infection (SSTI; N = 218,792), intestinal infectious diseases (IIDs; N = 218,792), urinary tract infection (N = 463,010), and central nervous system (CNS) infections (N = 218,792) among individuals of European ancestry. Independent genetic variants significantly (P < 10−8) associated with each exposure were considered instruments. The primary analysis used an inverse variance-weighted method, followed by a series of sensitivity analyses. Genetically predicted sleep disorders were associated with an increased risk of SSTI (odds ratio [OR], 1.29 [95% confidence interval (CI), 1.05–1.59]; P = .017). Genetically predicted depression was linked with an increased risk of CNS infections (OR, 1.59 [95% CI, 1.00–2.53]; P = .049) and SSTI (1.24 [95% CI, 1.03–1.49]; P = .024). Genetically predicted anxiety was associated with IIDs (OR, 1.19 [95% CI, 1.03–1.37]; P = .017) and SSTI (OR, 1.21 [95% CI, 1.02–1.43]; P = .029). There was no significant causal evidence for genetic prediction of nervous feelings and unspecified mental disorders in IIDs, CNS infections, SSTI, pneumonia, or urinary tract infection. Sensitivity analyses showed that the above causal association estimates were robust. In this MR study, we demonstrated a causal relationship between sleep disorders, depression, anxiety, and the risk of infectious diseases. However, no evidence was found to support causality between nervous feelings, unspecified mental disorders, and the risk of infectious diseases.
- Open Access
- 18 July 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Accuracy of Large Language Models When Answering Clinical Research Questions: Systematic Review and Network Meta-Analysis (Preprint)
BACKGROUND Large language models (LLMs) have flourished and gradually become an important research and application direction in the medical field. However, due to the high degree of specialization, complexity, and specificity of medicine, which results in extremely high accuracy requirements, controversy remains about whether LLMs can be used in the medical field. More studies have evaluated the performance of various types of LLMs in medicine, but the conclusions are inconsistent. OBJECTIVE This study uses a network meta-analysis (NMA) to assess the accuracy of LLMs when answering clinical research questions to provide high-level evidence-based evidence for its future development and application in the medical field. METHODS In this systematic review and NMA, we searched PubMed, Embase, Web of Science, and Scopus from inception until October 14, 2024. Studies on the accuracy of LLMs when answering clinical research questions were included and screened by reading published reports. The systematic review and NMA were conducted to compare the accuracy of different LLMs when answering clinical research questions, including objective questions, open-ended questions, top 1 diagnosis, top 3 diagnosis, top 5 diagnosis, and triage and classification. The NMA was performed using Bayesian frequency theory methods. Indirect intercomparisons between programs were performed using a grading scale. A larger surface under the cumulative ranking curve (SUCRA) value indicates a higher ranking of the corresponding LLM accuracy. RESULTS The systematic review and NMA examined 168 articles encompassing 35,896 questions and 3063 clinical cases. Of the 168 studies, 40 (23.8%) were considered to have a low risk of bias, 128 (76.2%) had a moderate risk, and none were rated as having a high risk. ChatGPT-4o (SUCRA=0.9207) demonstrated strong performance in terms of accuracy for objective questions, followed by Aeyeconsult (SUCRA=0.9187) and ChatGPT-4 (SUCRA=0.8087). ChatGPT-4 (SUCRA=0.8708) excelled at answering open-ended questions. In terms of accuracy for top 1 diagnosis and top 3 diagnosis of clinical cases, human experts (SUCRA=0.9001 and SUCRA=0.7126, respectively) ranked the highest, while Claude 3 Opus (SUCRA=0.9672) performed well at the top 5 diagnosis. Gemini (SUCRA=0.9649) had the highest rated SUCRA value for accuracy in the area of triage and classification. CONCLUSIONS Our study indicates that ChatGPT-4o has an advantage when answering objective questions. For open-ended questions, ChatGPT-4 may be more credible. Humans are more accurate at the top 1 diagnosis and top 3 diagnosis. Claude 3 Opus performs better at the top 5 diagnosis, while for triage and classification, Gemini is more advantageous. This analysis offers valuable insights for clinicians and medical practitioners, empowering them to effectively leverage LLMs for improved decision-making in learning, diagnosis, and management of various clinical scenarios. CLINICALTRIAL PROSPERO CRD42024558245; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024558245
- 18 July 2024
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessAccuracy of Large Language Models When Answering Clinical Research Questions: Systematic Review and Network Meta-Analysis (Preprint)
BACKGROUND Large language models (LLMs) have flourished and gradually become an important research and application direction in the medical field. However, due to the high degree of specialization, complexity, and specificity of medicine, which results in extremely high accuracy requirements, controversy remains about whether LLMs can be used in the medical field. More studies have evaluated the performance of various types of LLMs in medicine, but the conclusions are inconsistent. OBJECTIVE This study uses a network meta-analysis (NMA) to assess the accuracy of LLMs when answering clinical research questions to provide high-level evidence-based evidence for its future development and application in the medical field. METHODS In this systematic review and NMA, we searched PubMed, Embase, Web of Science, and Scopus from inception until October 14, 2024. Studies on the accuracy of LLMs when answering clinical research questions were included and screened by reading published reports. The systematic review and NMA were conducted to compare the accuracy of different LLMs when answering clinical research questions, including objective questions, open-ended questions, top 1 diagnosis, top 3 diagnosis, top 5 diagnosis, and triage and classification. The NMA was performed using Bayesian frequency theory methods. Indirect intercomparisons between programs were performed using a grading scale. A larger surface under the cumulative ranking curve (SUCRA) value indicates a higher ranking of the corresponding LLM accuracy. RESULTS The systematic review and NMA examined 168 articles encompassing 35,896 questions and 3063 clinical cases. Of the 168 studies, 40 (23.8%) were considered to have a low risk of bias, 128 (76.2%) had a moderate risk, and none were rated as having a high risk. ChatGPT-4o (SUCRA=0.9207) demonstrated strong performance in terms of accuracy for objective questions, followed by Aeyeconsult (SUCRA=0.9187) and ChatGPT-4 (SUCRA=0.8087). ChatGPT-4 (SUCRA=0.8708) excelled at answering open-ended questions. In terms of accuracy for top 1 diagnosis and top 3 diagnosis of clinical cases, human experts (SUCRA=0.9001 and SUCRA=0.7126, respectively) ranked the highest, while Claude 3 Opus (SUCRA=0.9672) performed well at the top 5 diagnosis. Gemini (SUCRA=0.9649) had the highest rated SUCRA value for accuracy in the area of triage and classification. CONCLUSIONS Our study indicates that ChatGPT-4o has an advantage when answering objective questions. For open-ended questions, ChatGPT-4 may be more credible. Humans are more accurate at the top 1 diagnosis and top 3 diagnosis. Claude 3 Opus performs better at the top 5 diagnosis, while for triage and classification, Gemini is more advantageous. This analysis offers valuable insights for clinicians and medical practitioners, empowering them to effectively leverage LLMs for improved decision-making in learning, diagnosis, and management of various clinical scenarios. CLINICALTRIAL PROSPERO CRD42024558245; https://www.crd.york.ac.uk/PROSPERO/view/CRD42024558245