Large language models (LLMs) such as ChatGPT, Claude, Llama, and Qwen are emerging as transformative technologies for the diagnosis and treatment of various diseases. With their exceptional long-context reasoning capabilities, LLMs are proficient in clinically relevant tasks, particularly in medical text analysis and interactive dialogue. They can enhance diagnostic accuracy by processing vast amounts of patient data and medical literature and have demonstrated their utility in diagnosing common diseases and facilitating the identification of rare diseases by recognizing subtle patterns in symptoms and test results. Building on their image-recognition abilities, multimodal LLMs (MLLMs) show promising potential for diagnosis based on radiography, chest computed tomography (CT), electrocardiography (ECG), and common pathological images. These models can also assist in treatment planning by suggesting evidence-based interventions and improving clinical decision support systems through integrated analysis of patient records. Despite these promising developments, significant challenges persist regarding the use of LLMs in medicine, including concerns regarding algorithmic bias, the potential for hallucinations, and the need for rigorous clinical validation. Ethical considerations also underscore the importance of maintaining the function of supervision in clinical practice. This paper highlights the rapid advancements in research on the diagnostic and therapeutic applications of LLMs across different medical disciplines and emphasizes the importance of policymaking, ethical supervision, and multidisciplinary collaboration in promoting more effective and safer clinical applications of LLMs. Future directions include the integration of proprietary clinical knowledge, the investigation of open-source and customized models, and the evaluation of real-time effects in clinical diagnosis and treatment practices.
Humans ; Large Language Models ; Tomography, X-Ray Computed

Objective To develop models for predicting response to chemotherapy combined with immunotherapy in patients with esophageal squamous carcinoma with various machine learning algorithms,and then select the optimal model.Methods A retrospective study was performed for 174 patients with esophageal squamous cell carcinoma undergoing chemotherapy combined with immunotherapy admitted in Department of Thoracic Surgery of the First Affiliated Hospital of Army Medical University from January 2022 to December 2023.The CT scans and clinical information were collected before treatment.They were randomly divided into a training set(n=122)and a testing set(n=52)in a ratio of 7∶3.CT radiomic features were extracted and selected,and then 5 machine-learning algorithms were employed to establish the prediction models,including radiomics model and clinical-radiomics model.Five-fold cross-validation was conducted on the training set,and the performance of the prediction models was evaluated on the testing set using receiver operating characteristic(ROC)curve and the F1 score.The best-performing model was further explained using local interpretable model-agnostic explanations(LIME)algorithm.Results Among the 174 patients,115(66.1%)achieved clinical remission.From the clinical information and CT images,1 clinical features and 10 radiomic features were identified.The area under of ROC curve(AUC)for the radiomics and clinical-radiomics models was 0.750(95%CI:0.616～0.883),and 0.766(95%CI:0.637～0.895),respectively.The F1 score of the optimal clinical-radiomics model was 0.829.LIME algorithm indicated that this best model demonstrated reliability in predicting individual samples.Conclusion The clinical-radiomics prediction model based on machine learning algorithm performs well,and can provide a reference for doctors'clinical decision-making by predicting the response to chemotherapy combined with immunotherapy in patients with esophageal squamous cell carcinoma.

AIM: To analyze the research status and future development trends of intermittent exotropia(IXT)by bibliometric study.METHODS: Bibliometrics methods were used and the related literatures in the Web of Science Core Collection(WoSCC)database from 2003 to 2022 were retrieved. CiteSpace6.2.R2 software was used to conduct visualized analysis of publications of one year, countries, institutions, journals, authors, references and keywords.RESULTS: A total of 620 literatures on IXT were retrieved from 2003 to 2022, and there has been a progressive increase in the number of publications. South Korea and the United States, Mayo Clinc and Holmes JM were the most productive and impactful country, institution and author, respectively. The Journal of American Association for Pediatric Ophthalmology and Strabismus(J AAPOS)published the most manuscripts(78 publications). The keywords with the strongest citation burst were surgery, epidemiology, alignment and recurrence.CONCLUSION: Visualized analysis conducted by CiteSpace software could objectively show the quantity changes and distribution of literatures on IXT from 2003 to 2022. Furthermore, the research hotspot of IXT has gradually shifted from surgery and epidemiology to fusion and recurrence.

Objective To investigate the relationship between serum levels of nuclear factor E2-related fac-tor 2(Nrf2),advanced oxidation protein products(AOPP)and blood lipid,liver fibrosis in patients with non-alcoholic steatohepatitis(NASH).Methods A total of 104 patients with NASH in Bishan Hospital Affiliated to Chongqing Medical University were selected as the study group,and 90 healthy people were selected as the control group.The serum levels of Nrf2 and AOPP in each group were detected and compared.Pearson or Spearman correlation analysis was used to analyze the relationship between serum Nrf2,AOPP levels and blood lipid,liver fibrosis in patients with NASH.Receiver operating characteristic(ROC)curve was used to e-valuate the diagnostic value of serum Nrf2,AOPP levels for NASH.Results The levels of triglyceride(TG),total cholesterol(TC),low density lipoprotein cholesterol(LDL-C),Nrf2 and AOPP in the study group were higher than those in the control group(P＜0.05),and the level of high density lipoprotein cholesterol(HDL-C)was significantly lower than that in the control group(P＜0.05).The serum levels of Nrf2 and AOPP in severe group were higher than those in moderate group and mild group(P＜0.05),and the serum levels of Nrf2 and AOPP in moderate group were higher than those in mild group(P＜0.05).Correlation analysis showed that serum Nrf2 and AOPP levels in NASH patients were positively correlated with TG,TC,LDL-C and the degree of liver fibrosis(P＜0.05),and negatively correlated with HDL-C(P＜0.05).The area under the curve(AUC)of serum Nrf2 for NASH diagnosis was 0.830(95％CI 0.780-0.880).The AUC of serum AOPP in the diagnosis of NASH was 0.866(95％CI 0.816-0.916).The AUC of the combined diagnosis of NASH was 0.925(95％CI 0.875-0.975).Conclusion The serum levels of Nrf2 and AOPP are increased in NASH patients,and they are closely related to blood lipids and liver fibrosis,which are expected to be effective indicators for the diagnosis of NASH.

Objective To evaluate the expression differences of C-reactive protein(CRP),procalcitonin(PCT)and cytokines[interferon-γ(IFN-γ),tumor necrosis factor-α(TNF-α),interleukin(IL)-10,IL-6,IL-4,IL-2,IL-17 A,and IFN-γ/IL-10]in human immunodeficiency virus(HIV)infection,tuberculosis(TB)infec-tion and HIV/TB co-infection in order to provide evidence for early clinical diagnosis of HIV/TB co-infection.Methods Medical records and clinical information of 103 patients with HIV(HIV group),42 patients with TB(TB group)and 67 patients with HIV/TB co-infection(HIV/TB group)were collected from January 2021 to September 2022.CRP,PCT and cytokines were detected in all patients,and their expression differ-ences among the three groups were analyzed.Receiver operating characteristic(ROC)curves were drawn and screening ability was evaluated.Results Compared with HIV group and TB group,CRP,PCT,IFN-γ,IL-10,IL-6 levels and IFN-γ/IL-10 in HIV/TB group were increased,with statistical significance(P＜0.001).Bon-ferroni correction results showed that CRP,PCT,IFN-γ,IL-6 levels and IFN-γ/IL-10 were significantly differ-ent between HIV and HIV/TB groups(all P＜0.001).ROC curve analysis showed that the area under the curve of CRP,PCT,IFN-γ,IL-6 and IFN-γ/IL-10 combined detection was 0.826,which was higher than that of single detection.Conclusion CRP,PCT,IFN-γ,IL-6,IFN-γ/IL-10 combined detection can help the early screening of HIV infection and HIV/TB co-infection.

Objective:To retrieve and summarize the evidence for postoperative observation time and bed rest time in patients undergoing percutaneous renal biopsy, so as to provide evidence-based basis for postoperative nursing of patients undergoing percutaneous renal biopsy.Methods:Using search terms such as "renal biopsy" and "renal puncture", based on the "6S" evidence resource pyramid model, evidence on postoperative observation time and bed rest time of percutaneous renal biopsy patients, including guidelines, best practice information books, evidence summaries, systematic reviews, expert consensus, and original research, was systematically searched on various guideline websites, evidence-based databases, original research databases, and professional association websites both domestically and internationally. The search period was from database establishment to July 1, 2023. After independent evaluation of the included literature by two researchers, evidence was extracted and summarized from the literature that met the quality standards.Results:A total of nine articles were included, including one guideline, five cohort studies, and three cross-sectional studies. This study summarized six best pieces of evidence from two aspects, namely postoperative observation time and postoperative bed rest time. For patients with low risk of complications after percutaneous renal biopsy, it was recommended to observe for 6 to 8 hours after surgery, but further shortening the observation time to 4 to 5 hours may also be safe and sufficient. For patients with high risk of complications after percutaneous renal biopsy, those living far from the hospital, those living alone, or those who may experience significant psychological stress after biopsy, it was recommended to extend the observation time to up to 24 hours.Conclusions:Based on existing evidence, it seems feasible to shorten the bed rest time after percutaneous renal biopsy. A large amount of high-quality research is still needed to explore the optimal postoperative bed rest time.

Objective:To investigate the level of beliefs about medication in young patients with newly treated pulmonary tuberculosis and analyze its influencing factors.Methods:This was a cross-sectional study. Using the convenient sampling method, a total of 320 young newly treated pulmonary tuberculosis patients who visited the designated tuberculosis hospitals in Kashgar Prefecture and Hotan Prefecture of Xinjiang Uygur Autonomous Region from January 2022 to April 2023 were selected as the research objects. The investigation was carried out with the General Information Questionnaire, Beliefs about Medicines Questionaire Specific (BMQ-S), 8-Item Morisky Medication Adherence Scale (MMAS-8), Brief Illness Perception Questionnaire (BIPQ) and Tuberculosis-related Stigma Scale (TSS). Multiple linear regression was used to analyze the influencing factors of medication beliefs in young patients with newly treated tuberculosis.Results:A total of 320 questionnaires were distributed in this study, and 302 valid questionnaires were collected, with an effective response rate of 94.38% (302/320). The total score of BMQ-S of 302 young patients with newly treated tuberculosis was -1.00 (-2.00, 1.00), MMAS-8 score was 5.38 (4.75, 5.75), BIPQ score was 37.00 (24.00, 44.00) and TSS score was 48.00 (44.00, 52.00). The results of multiple linear regression analysis showed that comorbidities, medication adherence, disease perception and stigma were the influencing factors of medication beliefs in young newly treated pulmonary tuberculosis patients ( P<0.01) . Conclusions:The medication belief level of young patients with newly treated pulmonary tuberculosis needs to be improved. Medical staff should correct the negative cognition of the patient's disease, emphasize the benefits of drug treatment and enhance the patients' beliefs in the necessity of medication.

Objective:To investigate the brain functional connectivity and network of patients with bipolar depression (BD) during emotion recognition, and to assess the correlation of brain functional indicators with clinical data.Methods:From August 2023 to September 2024, a total of 34 BD patients and 35 healthy controls were recruited for the study. The 17-item Hamilton depression scale(HAMD-17) and the Hamilton anxiety scale (HAMA) were used to assess the depression and anxiety symptoms of BD patients. While all the participants performed the emotion recognition task, a functional near-infrared spectroscopy (fNIRS) with 52 channels was utilized to collect their oxyhemoglobin concentrations in dorsolateral prefrontal cortex (DLPFC), medial prefrontal cortex (mPFC), and temporal lobe (TL). Brain functional indicators, including brain functional connectivity(BFC) and brain functional network (BFN) parameters, were extracted using fNIRS analysis algorithms. SPSS 26.0 software was used to perform t-test, Wilcoxon rank sum test, chi-square test, Pearson correlation and multiple linear regression analysis. Results:The emotion recognition accuracy of BD patients ((69.40±18.32) %) was lower than that of healthy controls ((76.94±6.09)%)( t=-2.13, P=0.036), and the average spending time ((4.00±1.78) s) of BD patients was longer than that of healthy controls ((2.69±1.27) s) ( t=-2.81, P=0.007). The BFC values of right DLPFC-right mPFC, right DLPFC-left TL, right DLPFC-left mPFC, right TL-right mPFC, right TL-left TL, right TL-left mPFC, right mPFC-left DLPFC, and right mPFC-left mPFC, the clustering coefficient, and small-world coefficient of BD patients were lower than those of healthy controls (all P<0.05).The characteristic path length of BD patients was higher than that of healthy controls ( t=1.27, P<0.05). For BD patients, the total score of HAMD-17 were significantly and negatively correlated with the clustering coefficient, small-world coefficients, and BFC values of right DLPFC-left mPFC and right mPFC-left DLPFC ( r=-0.45--0.27, all P<0.05).The total score of HAMA was significantly and positively correlated with the BFC value of right DLPFC-left mPFC ( r=0.30, P<0.05). The results of multiple linear regression analysis showed that the clustering coefficient ( β=-0.12, t=-0.69, P=0.009), small-world coefficients ( β=-0.22, t=-1.05, P=0.013), and BFC values of right DLPFC-left mPFC ( β=-0.47, t=-2.29, P=0.030) and right mPFC-left DLPFC( β=-0.20, t=-0.90, P=0.010) were influencing factors of depressive symptoms in BD patients. The BFC value of right DLPFC -left mPFC ( β=0.37, t=2.36, P=0.026) was influencing factor for anxiety symptoms in BD patients. Conclusions:There are abnormalities in some BFC values and BFN parameters in the DLPFC, mPFC, and TL brain regions during the emotional recognition task, which are related to the clinical symptoms. The abnormalities may be associated with the emotional recognition deficits in BD patients. These findings may provide theoretical guidance for the development of clinical intervention strategies for BD patients.

Objective To understand the change trend and characteristics of nosocomial infection through the comprehensive surveillance on nosocomial infection in infectious diseases specialized hospitals dur-ing 2017-2023 to provide an evidence for the prevention,control and management of nosocomial infection.Methods The surveillance indicators of nosocomial infection in a hospital during 2017-2023 were collected.The nosocomial infection rate,nosocomial infection rate in different inpatient wards,nosocomial infection sites,nosocomial infection pathogenic bacterial distribution and susceptibility factors conducted the statistical analysis.Results A total of 93 254 patients were admitted and treated during 2017-2023.The nosocomial in-fection rate and infection case-times rate showed the decreasing trend(P＜0.05).The case-times rate of the patients in the AIDS wards was 3.75％,which was higher than 0.79％in the tuberculosis wards(P＜0.05).The nosocomial infection case-times rate in the two wards areas during 2017-2023 showed the decreasing trend(P＜0.05).The infection sites were mainly the respiratory system(61.56％),blood system(9.44％)and urinary system(8.61％).A total of 803 strains of pathogenic bacteria were detected out,which were mainly Gram negative bacteria(63.89％),the top five were in turn Klebsiella pneumoniae(16.19％),Esche-richia coli(15.57％),Acinetobacter baumannii(10.83％),Pseudomonas aeruginosa(8.84％)and Staphylo-coccus aureus(7.22％).The top three of susceptibility factors were low immune function(58.78％),long term antibiotic use(11.29％)and ventilator use(9.20％).Conclusion Initiatively carrying the hospital infec-tion surveillance could accurately grasp the incidence trends and provide the direction and data support for pre-vention and control priorities.

Objective: To analyze the trend and characteristics of school tuberculosis epidemic in Shijiazhuang from 2011 to 2020, and to provide reference for school tuberculosis prevention and control. Methods: Descriptive methods were used to analyze the epidemiological characteristics of tuberculosis among students and the epidemic situation of tuberculosis in schools in Shijiazhuang from 2011 to 2020. The χ 2 test and χ 2 trend test were used to analyze the characteristics and trend of school tuberculosis. Results: A total of 4 896 cases of tuberculosis were registered among students in Shijiazhuang from 2011 to 2020. The average annual registered incidence rate of students was 24.69/100 000, and the difference in incidence rate was statistically significant ( χ 2=318.50, P <0.01) the overall registered incidence rate of tuberculosis among students in the past 10 years was on the rise ( χ 2 trend =87.79, P <0.01). Among the student cases, male accounted for 53.89%, female accounted for 46.11%. The age group of students aged >18 and above accounted for the largest proportion(50.35%), followed by the age group aged 16-18( 35.80 %). Most students cases occurred in April and September-November, with September the highest(12.03%). A total of 22 clustered outbreaks (174 cases) and the aggregate epidemic accounted for 3.55% of the total number of students with tuberculosis. Conclusion The prevention and control of tuberculosis epidemic in schools in Shijiazhuang should not be underestimated, and strengthen the supervision, management, publicity and education of students in key age groups to avoid clusters of outbreaks.