Background: The application of large language models (LLMs) in medicine has received increasing attention, showing significant potential in teaching, research, and clinical practice, especially in knowledge extraction, management, and understanding. However, the use of LLMs in Traditional Chinese Medicine (TCM) has not been thoroughly studied. This study aims to provide a comprehensive overview of the status and challenges of LLM applications in TCM.
Methods: A systematic search of five electronic databases and Google Scholar was conducted between November 2022 and April 2024, using the Arksey and O'Malley five-stage framework to identify relevant studies. Data from eligible studies were comprehensively extracted and organized to describe LLM applications in TCM and assess their performance accuracy.
Results: A total of 29 studies were identified: 24 peer-reviewed articles, 1 review, and 4 preprints. Two core application areas were found: the extraction, management, and understanding of TCM knowledge, and assisted diagnosis and treatment. LLMs developed specifically for TCM achieved 70% accuracy in the TCM Practitioner Exam, while general-purpose Chinese LLMs achieved 60% accuracy. Common international LLMs did not pass the exam. Models like EpidemicCHAT and MedChatZH, trained on customized TCM corpora, outperformed general LLMs in TCM consultation.
Conclusion: Despite their potential, LLMs in TCM face challenges such as data quality and security issues, the specificity and complexity of TCM data, and the nonquantitative nature of TCM diagnosis and treatment. Future efforts should focus on interdisciplinary talent cultivation, enhanced data standardization and protection, and exploring LLM potential in multimodal interaction and intelligent diagnosis and treatment.
Topical nonsteroidal anti-inflammatory drugs (NSAIDs) are commonly used to treat sports injuries, but evidence-based medical guidance for their standardized and rational use is lacking. This guideline working group identified clinically important issues, obtained the full opinions of patients and clinical staff, and discussed them with the expert group. Based on evidence from the literature, the “clinical practice guidelines for topical NSAIDs in the treatment of sports injuries” were formulated following the methods and principles of international guidelines. According to these guidelines, 7 clinical concerns were ultimately selected, and 22 recommendationswere formed. These included the status, indications, contraindications, efficacy, combined application, use in special populations, adverse reactions, and countermeasures of topical NSAIDs in the treatment of sports injuries. The purpose of these guidelines is to provide evidence-based recommendations for practitioners in the fields of orthopedics, sports medicine, rehabilitationmedicine, and sports science, as well as other fields, in the treatment of sports injuries to promote more standardized and rational use of topical NSAIDs.
Objective: Current QI reports within the literature frequently fail to provide enough information regarding interventions, and a significant number of publications do not mention the utilization of a guiding model or framework. The objective of this scoping review was to synthesize the characteristics of hospital-based QI interventions and assess their alignment with recommended quality goals.
Methods: This scoping review followed the JBI methodology for scoping reviews to synthesize existing literature on hospitalbased QI interventions and reporting using the PRISMA Extension for scoping reviews. Included studies involved a hospital-based QI intervention that was evaluated through the Development of the Quality Improvement Minimum Quality Criteria Set (QIMQCS) framework, reporting on hospital users’ (i.e., practitioners and patients) data. We searched Medline, CINAHL, Embase and PubMed databases for primary research published between 2015 and 2024. Grey literature was also examined. A narrative synthesis guided the integration of findings.
Results: From 1398 identified records, 70 relevant records were included. Results indicate a wide variation in QI frameworks and methods used by the included studies. The QI interventions most frequently assessed were organizational-focused (n = 59), followed by professional-related interventions (n = 41) and patient-care interventions (n = 24). Thereweremultiple facilitators and barriers across organizational, professional, and patient care levels found in the included studies. Examples of facilitators were instrumental in driving successful QI initiatives included education, training, active leadership, and stakeholder engagement. Conversely, barriers such as time constraints, resource limitations, and resistance were highlighted.
Conclusion: Existing QI publications lack sufficient detail to replicate interventions. Using a model or framework to guide the conduct of a QI-activity may support a more robustly designed and well-conducted project. The variation of reporting characteristics suggests that future research should focus on the development of a pragmatic tool for use by front-line clinicians to support consistent and rigorous conduct of QI projects.
Objective: Extraversion is a fundamental personality dimension that contributes to an individual's overall health and well-being. Many studies have examined the neural bases of extraversion but these results are inconsistent. This study adopted a meta-analysis approach to examine the brain activity correlates of extraversion by incorporating functional neuroimaging studies in the context of positive affect/emotional stimuli.
Methods: A systematic literature search was performed in the databases of PubMed, Web of Science Core Collection, Embase, PsycInfo, CNKI, WanFang, and Weipu. The demographic characteristics and basic information of the included studies were first summarized. Then, a meta-analysis was conducted using anisotropic effect-size seed-based d mapping. Jackknife sensitivity analysis was next conducted to examine the reliability of the findings. Finally, meta-regression analysis was performed to test the potential demographic effects (i.e., sex and age) on the association between extraversion and brain activity.
Results: A total of 11 studies were included in the meta-analysis. The results of meta-analysis revealed robust and consistent positive correlations between extraversion and activation of right inferior frontal gyrus/insula, right angular gyrus, and left precentral gyrus during positive affect processing. In contrast, brain activity in the right striatum during positive affect processing was negatively associated with extraversion. Additionally, meta-regression analysis revealed sex as a moderator in the relationship between extraversion and right insular activation.
Conclusions: Overall, our study is the first to provide a comprehensive understanding of functional brain activation patterns of extraversion, which may be helpful for targeting of specific brain regions in personalized interventions for extraversion-related psychological/physical illness.
Objective: This study aimed to investigate the relationship between recurrent aphthous ulcers (RAU) and dysbiosis of the oral microbiota.
Methods: A systematic search was conducted across several databases, including the Cochrane Library, PubMed, Web of Science Core Collection, Embase, Scopus, CBM, CNKI, WanFang, and VIP, to identify relevant studies from the inception of the library until September 20, 2024. All included studies were evaluated for quality using the Newcastle–Ottawa scale. These studies assessed the diversity and abundance of oral microorganisms in patients with RAU in comparison to healthy individuals. Results were synthesized via random-effects meta-analysis (I2 statistic for heterogeneity). Mean differences (MD) and 95% confidence intervals (CI) were applied to evaluate the pooled effects of outcomes. The systematic review was registered in PROSPERO (CRD42024615516).
Results: A total of 10 studies involving 343 patients and 348 controls were included in this study, 7 of these were utilized for meta-analysis to conduct a quantitative assessment. The meta-analysis showed that in terms of alpha diversity, saliva samples collected from patients with RAU exhibited a significantly lower Chao 1 index of alpha diversity (MD = –41.22, 95% CI –64.34 to –18.09,p < 0.01), and that patients with ulcerated sites in the oral mucosa demonstrated a higher Chao 1 index when compared to healthy sites within the oral mucosa (MD = 27.48, 95% CI 2.98 to 51.97,p = 0.03).
Conclusions: Significant differences in the Chao 1 index between patients and healthy controls, along with variations in beta diversity and the relative abundance of taxa, indicate that microbial dysbiosis may play a crucial role in the development of RAU.
Objective: With a steadily rising prevalence, nonalcoholic fatty liver disease (NAFLD) was a leading global cause of liver-related health problems. In the clinical management of NAFLD, various western pharmaceuticals were widely utilized. This network meta-analysis aimed to evaluate the effectiveness of common western medications for NAFLD patients.
Methods: We systematically reviewed and screened articles based on predesigned criterion about western medications for NAFLD, which were from Embase, Cochrane Library, PubMed, CNKI, WanFang, and China Science and Technology Journal Database until August 1, 2024. Eligible studies included randomized controlled trials of patients aged 18 or older with NAFLD, comparing Western medicines to placebos or other Western medicine treatments. The risk of bias assessment tool 2.0 from the Cochrane system was used to assess the quality of the included articles. A Bayesian network meta-analysis was conducted using WinBUGS 1.4.3 with a random-effectsmodel and Markov Chain Monte Carlo methods. Treatment rankingswere based on Surface Under the Cumulative Ranking Curve (SUCRA) values, and heterogeneity was assessed with I2 and Q statistics. The outcomes were analyzed in WinBUGS and visualized using Stata 14.0, generating network plots and cumulative probability rankings to compare treatment effects. The systematic review was registered in PROSPERO (CRD42024509176).
Results: Based on 37 included articles involving 7673 patients, pioglitazone demonstrated the most significant effects in resolving nonalcoholic steatohepatitis without worsening fibrosis, increasing high-density lipoprotein cholesterol levels, and achieving a ≥ 2-point reduction in NAFLD activity scores (odds ratio [OR] = 0.09, 95% confidence interval [CI]: 0.01 to 0.81), with a SUCRA probability of 91.4%. Aldafermin showed remarkable effects in improving liver function markers, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and γ-glutamyl transpeptidase, with cumulative probabilities of 90% for ALT and 69.8% for AST. Cluster analysis revealed that Resmetirom and Aldafermin were superior options for enhancing liver function, while pioglitazone emerged as the best treatment for the comprehensive improvement of NAFLD.
Conclusions: Pioglitazone outperformed other western medicines in terms of overall efficacy when treating NAFLD, but Aldafermin and Resmetirom showed superior improvement in liver function. This study provided a certain level of support for the use of specific clinical medications.
Intelligent traditional Chinese medicine (TCM) is a key pathway toward the modernization and globalization of TCM in the era of artificial intelligence. Due to its unique terminology and diagnostic framework, TCM's intelligentization process has long faced a range of challenges, from the digitization and formalization of knowledge bases to the differentiation of syndromes and personalized treatment. Recently, the advent of large language models (LLMs) like GPTs has marked a transformative milestone in semantic understanding tasks, attracting widespread attention from the medical, academic, and industrial communities. Nonetheless, LLMs often suffer from accuracy and logical reasoning limitations within specific fields and may manifest hallucinations in the generative outputs. Through a comprehensive review of existing literature and empirical analyses, this study delves into the potential and challenges of adapting LLMs to TCM. Promising perspectives on future developments at this innovative intersection are discussed.
Objective: Leukemia is a type of blood cancer that begins in the bone marrow and results in high numbers of abnormal white blood cells. Automated detection and classification of leukemia and its subtypes using artificial intelligence (AI) and machine learning (ML) algorithms plays a significant role in the early diagnosis and treatment of this fatal disease. This study aimed to review and synthesize research findings on AI-based approaches in leukemia detection and classification from peripheral blood smear images.
Methods: A systematic literature search was conducted across four e-databases (Web of Science, PubMed, Scopus, and IEEE Xplore) from January 2015 to March 2023 by searching the keywords “Leukemia, ” “Machine Learning, ” and “Blood Smear Image, ” as well as their synonyms. All original journal articles and conference papers that used ML algorithms in detecting and classifying leukemia were included. The study quality was assessed using the Qiao Quality Assessment tool.
Results: From 1325 articles identified through a systematic search, 190 studies were eligible for this review. The mean validation accuracy (ACC) of the ML methods applied in the reviewed studies was 95.38%. Among different ML methods, modern techniques were mostly considered to detect and classify leukemia (60.53% of studies). Supervised learning was the dominant ML paradigm (79% of studies). Studies utilized common ML methodologies for leukemia detection and classification, including preprocessing, feature extraction, feature selection, and classification. Deep learning (DL) techniques, especially convolutional neural networks, were the most widely used modern algorithms in the mentioned methodologies. Most studies relied on internal validation (87%). Moreover, K-fold cross-validation and train/test split were the commonly employed validation strategies.
Conclusion: AI-based algorithms are widely used in detecting and classifying leukemia with remarkable performance. Future studies should prioritize rigorous external validation to evaluate generalizability.
Objective: To assess the quality of harm reporting in randomized controlled trials (RCTs) published in high-impact general medical journals.
Study Design and Setting: Publications of RCTs involving drugs compared with placebo controls, that were published in five general medical journals with high Impact Factors were identified from January 2022 to December 2023. Data relating to the presentation and discussion of harm were extracted and analyzed based on the Consort Harm framework.
Results: We identified 175 eligible RCTs (AIM: n = 5; BMJ: n = 8; JAMA: n = 26,Lancet: n = 64, and NEJM: n = 72). None of the studies referenced the CONSORT Harms 2004 statement. Seventy-one percent of studies (n = 125) did not mention how harm data about patients’ symptoms were collected and 86.3% of the analyses (n = 151) were limited to descriptive statistics. Only 45.1% of studies (n = 79) discussed the balance of benefits and harms. Common limitations included unclear methodological details, selective reporting, and inadequate analysis of results.
Conclusions: RCTs published in five highly cited general medical journals contain deficiencies in harm reporting. The recently updated Consort Harm 2022 provides an implementable evaluation and guidance tool and should be actively promoted among researchers, reviewers, and journal editors. More attention to adequate and reasonable reporting requirements for harms in RCTs is necessary to provide a better opportunity for evidence-based decision making.
Aim: To provide a comprehensive assessment of the efficacy and safety of pharmacological interventions for AP.
Methods: This was an overview of systematic reviews based on randomized controlled trials comparing pharmacological interventions with placebo or blank control in adults with AP. We searched PubMed, Embase, and the Cochrane Database of Systematic Reviews from inception to January 13, 2024, with an update on February 4, 2025. The effect value of each medication on each outcome of interest defined as a “combo” was assessed. Findings were categorized as efficacious, not efficacious, or inconclusive.
Results: Fifteen reviews (167 unique trials, 12, 930 participants) reported the efficacy of 14 medications on 5 outcomes, yielding 35 distinct combos. Seven combos showed efficacy with low certainty evidence: low molecular weight heparin (risk ratio 0.31, 95% confidence interval, 0.18–0.51), omega-3 fatty acids (0.30, 0.14–0.65), and antioxidants (0.69, 0.49–0.98) for mortality; low molecular weight heparin (0.38, 0.22–0.65), chengqi-series decoctions (0.48, 0.36–0.63), and ulinastatin (0.43, 0.24–0.78) for multiple organ failure; and neostigmine (mean difference –2.81, 95 % confidence interval –3.75 to –1.87) for length of intensive care unit stay. Half of the remaining combos showed no efficacy, while the other half was inconclusive for very low certainty evidence. Safety data were limited, with one review reporting no significant adverse events for neostigmine.
Conclusions: Some pharmacological interventions exhibited potential efficacy for specific AP outcomes, albeit with low certainty evidence. Further verifying those medications is crucial in advancing the treatment landscape for AP.
Aim: Digital health technology in swallowing rehabilitation offers personalized exercises, remote monitoring, and real-time feedback, enhancing accessibility and effectiveness of therapy. This scoping review was conducted to summarize what types and features of digital technology-based dysphagia rehabilitation interventions exist, how they are applied in patients with dysphagia, and what the effectiveness and facilitators and barriers to intervention application are.
Methods: We searched Medline Complete, Embase, CINAHL, Scopus, and gray literature for articles published between January 2000 and June 2023. We used subheadings and terms related to digital health, dysphagia, and rehabilitation to search for articles. The included studies were mapped according to the types and features, effectiveness, enablers, barriers, and future improvements of swallowing rehabilitation using digital technologies.
Results: Twenty-five studies met the inclusion criteria. Three types of digital swallowing rehabilitation interventions were identified: home-based rehabilitation using the mHealth app, synchronous telepractice and monitoring, as well as game-based biofeedback and tracking. The included studies reported positive results regarding physiological changes in swallowing function, swallowing performance, and quality of life. Digital unfamiliarity, resources for digital access, and technical issues related to the failure of the mobile device operating system were identified as barriers to the use of digital swallowing rehabilitation technology and future improvements.
Conclusions: Digital technology has potential value in dysphagia rehabilitation. In the future, developing various interventions utilizing the advantages of digital technology and conducting additional research to validate their effectiveness is necessary. Additionally, improved digital familiarity, better accessibility, better technology, and management practices will be needed.
Objective: To investigate and report the psychosocial health status from a survey of adolescent populations in Zigong City, Sichuan Province, China, and analyze the overall condition of adolescent mental health in economically less-developed prefecture-level urban areas of China.
Methods: A survey on the mental health status of 95, 455 adolescents in Zigong was conducted using eight scales: the Strengths and Difficulties Questionnaire for Students, the Pre-psychosis Questionnaire, the Patient Health Questionnaire, the Generalized Anxiety Disorder Scale, the Children's Revised Impact of Event Scale, the Pittsburgh Sleep Quality Index, the Internet Gaming Disorder Scale Short Form, and the Warwick-Edinburgh Mental Well-Being Scale. The survey assessed mental health across eight dimensions: emotional and behavioral problems, prodromal psychosis risk, depression, anxiety, stress reactions, sleep quality, internet addiction, and positive mental health. The statistical significance of the results was validated using the Kruskal–Wallis test, Mann–Whitney U test, and Chi-square test.
Results: Overall, 76.14% of adolescents in Zigong exhibited normal psychosocial health status, indicating no immediate cause for concern. Approximately 5.58% of students were identified as being in a state of extremely high-risk psychosocial alertness, necessitating prompt intervention measures. Among all educational stages, middle-school students displayed the most severe psychosocial health issues, with approximately 8.39% classified as being in an extremely high-risk psychosocial alertness state. Furthermore, female adolescents exhibited more pronounced mental health issues compared with their male counterparts.
Conclusions: The prominence of psychological issues among Chinese adolescents in less-developed cities underscores the urgent need for normalized and systematic intervention measures. Particular attention should be directed toward implementing targeted interventions for middle-school students and female students to address their psychosocial well-being.
Objective: To identify and map systematic reviews on the effectiveness of treatment for depressive disorders among adults.
Methods: We retrieved systematic reviews and meta-analyses of randomized controlled trials involving adults with depressive symptoms from twelve English and four Chinese databases (June 21, 2022). Using an interactive map, we visualized the effectiveness of evidence on depression based on an intervention-outcome framework. The interventions included psychotherapy, pharmacotherapy, complementary and alternative treatments, and others. The outcomes included the remission of depressive symptoms, symptoms of depressive disorder, life and social skills, and adverse events.
Results: We included 994 systematic reviews and meta-analyses, including 32 that were review protocols, highlighting the distribution of psychotherapy, pharmacotherapy, and complementary and alternative treatments. However, the evidence and gap map (EGM) revealed significant gaps in evidence for specific interventions, populations, outcomes, and regions. While psychotherapy, pharmacotherapy, and complementary and alternative treatments dominate the landscape, the review highlighted a lack of research on interventions for specific types of depression, such as depression in people with bipolar disorder and treatment-resistant depression. It was a similar situation for underserved populations, including young and middle-aged adults, males, sexual minority individuals, and people with disabilities. The map also suggested the need for more research on the potential risks and side effects associated with both pharmacological and nonpharmacological treatments.
Conclusions: The contribution of this EGM was to present the available evidence on psychotherapy, pharmacotherapy, and complementary and alternative treatments for depression in adults, making available an evidence base that could inform future policy decisions and practice. It also identified evidence gaps in interventions, outcomes, population, regions, and evidence confidence. The need for further research on tailored treatments for specific populations was highlighted.
Objective: Cerebrospinal fluid (CSF) leakage may occur during transsphenoidal resection of pituitary adenomas, with implications for patient prognosis. However, although the risk factors for intraoperative CSF leakage have been widely studied and continuously explored, there are still some unknown factors that play a role.
Methods: A retrospective study was conducted by collecting clinical data from 281 patients who underwent transsphenoidal resection of pituitary adenoma at Nanjing Drum Tower Hospital between October 2020 and October 2022.
Results: The results showed CSF leakage occurred in 60 (21.4%) patients. In the univariate analysis, we found that tumor anteroposterior diameter (p = 0.024), hepatic insufficiency (p = 0.004), tough tumor texture (p < 0.001), and surgical protocol as complete removal (p = 0.01) significantly predicted the occurrence of CSF leakage. In the univariate analysis, we found that anteroposterior diameter was associated with the occurrence of intraoperative CSF leakage compared with another diameter (p = 0.04), indicating that the probability of CSF leakage significantly increased at the anteroposterior diameter of pituitary adenoma ≥2 cm. We included anteroposterior tumor diameter, hepatic insufficiency, complete removal, and tumor tough texture together in a multifactorial logistic regression analysis and produced a nomogram graph based on the results. In the multifactorial logistic regression analysis, the anteroposterior tumor diameter, tumor tough texture, hepatic insufficiency, and the surgical option of complete removal predicted intraoperative CSF leakage with an AUC of 0.804.
Conclusions: This provides a reference for assessing the risk of CSF leakage in patients before and early in the surgical procedure and may have clinical implications for better patient treatments.
Objective: Inappropriate polypharmacy increases the risk of medication-related issues. Adequate management of polypharmacy is a challenge involving different healthcare professionals, complex decision-making and ideally including patient involvement. The objective of this scoping review was to provide an overview of national recommendations for medication management of patients with polypharmacy in primary care.
Methods: A scoping review of clinical practice guidelines focusing on medication management in adults with polypharmacy, applicable to primary care was performed. Databases (G-I-N, Turning Research into Practice and PubMed), network, and a global report were screened for guidelines published after 2000 in English, Dutch, German, Spanish, French, or Russian. Raw data were extracted in duplicate using an extraction framework focusing on strategies, involvement of professionals, patient involvement, and implementation. Qualitative content analysis was used. Guideline quality was assessed using AGREE-II. The study was registered with the Open Science Framework.
Results: Eight guidelines originating from eight countries were included. The most common recommended strategy was a medication review conducted by a general practitioner and/or a community pharmacist. Tasks and target population differed per guideline. Most guidelines recommended involving the patient in the process, mostly to elicit the patient’s experiences and treatment goals. Few guidelines included advice on the implementation of recommendations. Three out of eight guidelines were of good quality (AGREE-II score >70% in 5/6 domains).
Conclusions: Most guidelines recommended a medication review, with patient involvement, as a strategy for medication management in polypharmacy in primary care. Guidance on task division and implementation of guidelines in practice was less clear. This review illustrates room for guideline improvements.
Objective: To evaluate the methodological and reporting quality of clinical practice guidelines/expert consensus for ambulatory surgery centers published since 2000, combining manual assessment with large language model (LLM) analysis, while exploring LLMs’ feasibility in quality evaluation.
Methods: We systematically searched Chinese/English databases and guideline repositories. Two researchers independently screened literature and extracted data. Quality assessments were conducted using AGREE II and RIGHT tools through both manual evaluation and GPT-4o modeling.
Results: 54 eligible documents were included. AGREE II domains showed mean compliance: Scope and purpose 25.00%, Stakeholder involvement 20.16%, Rigor of development 17.28%, Clarity of presentation 41.56%, Applicability 18.06%, Editorial independence 26.39%. RIGHT items averaged: Basic information 44.44%, Background 36.11%, Evidence 14.07%, Recommendations 34.66%, Review and quality assurance 3.70%, Funding and declaration and management of interests 24.54%, Other information 27.16%. LLMs’-evaluated documents demonstrated significantly higher scores than manual assessments in both tools. Subgroup analyses revealed superior quality in documents with evidence retrieval, conflict disclosure, funding support, and LLM integration (P < 0.05).
Conclusions: Current guidelines and consensus related to day surgery need to improve their methodological quality and quality of reporting. The study validates LLMs’ supplementary value in quality assessment while emphasizing the necessity of maintaining manual evaluation as the foundation.