With the rapid development of artificial intelligence technology,Large Language Models(LLMs)have in-creasingly been applied in the medical field,particularly demonstrating significant potential in the prevention and management of medical disputes,patient management,and medical quality control.This article,from the perspective of cooperative governance,summarizes and analyzes the current application of LLMs in the field of medical dispute from the perspectives of the governments,research and development companies,medical staff,patients and all other relevant parties.It delves into the advantages of LLMs in dispute prediction,document writing,quality control,patient management,and case retrieval,and systematically analyzes the challenges faced in data security,privacy protection,professional competence,ethical and legal issues,and personnel utilization.The aim of this article is to provide references and development suggestions for further research and practice of LLMs in the field of medical disputes.

With the rapid development of artificial intelligence technology,Large Language Models(LLMs)have in-creasingly been applied in the medical field,particularly demonstrating significant potential in the prevention and management of medical disputes,patient management,and medical quality control.This article,from the perspective of cooperative governance,summarizes and analyzes the current application of LLMs in the field of medical dispute from the perspectives of the governments,research and development companies,medical staff,patients and all other relevant parties.It delves into the advantages of LLMs in dispute prediction,document writing,quality control,patient management,and case retrieval,and systematically analyzes the challenges faced in data security,privacy protection,professional competence,ethical and legal issues,and personnel utilization.The aim of this article is to provide references and development suggestions for further research and practice of LLMs in the field of medical disputes.

Objective To investigate the application value of clinical-radiomics nomogram in predicting the expression of liver kinase B1(LKB1)in non-small cell lung cancer(NSCLC)before surgery.Methods A total of 140 NSCLC patients were randomized into training group(n=106)and validation group(n=34)according to the ratio of 7∶3.The training group was used as the study cohort to screen the clinically independent predictors and radiomics characteristics related to LKB1 expression,and the clinical model,radiomics model and clinical-radiomics nomogram model were constructed,respectively.The predictive performance of the three models was analyzed using the receiver operating characteristic(ROC)curve in the training group,and validated in the validation group.The calibration curve was used to assess the consistency between the predicted results of nomogram model and the actual observations,and the decision curve was used to evaluate the clinical benefit of the nomogram model.Results The clinical model consisted of pathological type and hilal/mediastinal lymphadenopathy,the radiomics model consisted of Radiomics score(Radscore),and the nomogram model consisted of Radscore,pathological type and hilal/mediastinal lymphadenopathy.In the training group,the area under the curve(AUC)of the nomogram model,radiomics model and clinical model was 0.884,0.843 and 0.788,respectively.In the validation group,the AUC of the three models were 0.976,0.851,and 0.912,respectively.The calibration curve analysis showed good consis-tency between the predicted results of nomogram model and the actual observations,and the decision curve showed that the model had good clinical benefit.Conclusion Radiomics combined with clinical risk factors can effectively predict the expression of LKB1 in NSCLC patients before surgery,so as to contribute to the formulation of therapeutic strategies in clinical practice.

Objective To investigate the application value of clinical-radiomics nomogram in predicting the expression of liver kinase B1(LKB1)in non-small cell lung cancer(NSCLC)before surgery.Methods A total of 140 NSCLC patients were randomized into training group(n=106)and validation group(n=34)according to the ratio of 7∶3.The training group was used as the study cohort to screen the clinically independent predictors and radiomics characteristics related to LKB1 expression,and the clinical model,radiomics model and clinical-radiomics nomogram model were constructed,respectively.The predictive performance of the three models was analyzed using the receiver operating characteristic(ROC)curve in the training group,and validated in the validation group.The calibration curve was used to assess the consistency between the predicted results of nomogram model and the actual observations,and the decision curve was used to evaluate the clinical benefit of the nomogram model.Results The clinical model consisted of pathological type and hilal/mediastinal lymphadenopathy,the radiomics model consisted of Radiomics score(Radscore),and the nomogram model consisted of Radscore,pathological type and hilal/mediastinal lymphadenopathy.In the training group,the area under the curve(AUC)of the nomogram model,radiomics model and clinical model was 0.884,0.843 and 0.788,respectively.In the validation group,the AUC of the three models were 0.976,0.851,and 0.912,respectively.The calibration curve analysis showed good consis-tency between the predicted results of nomogram model and the actual observations,and the decision curve showed that the model had good clinical benefit.Conclusion Radiomics combined with clinical risk factors can effectively predict the expression of LKB1 in NSCLC patients before surgery,so as to contribute to the formulation of therapeutic strategies in clinical practice.

Goiter is a kind of non-inflammatory and non-neoplastic hyperplasia and enlargement. Many studies have shown that substances such as thiocyanates and isothiocyanates can prevent the development of a variety of tumors. However, some studies have also found that such substances can lead to goiter. In this article, relevant information on common goitrogen in food are collected to explore their mechanism of action, laying a foundation for guiding residents to maintain a healthy and balanced diet.

Objective:To study the diagnostic value of artificial intelligence B-ultrasound image computer-aided diagnosis system (hereinafter referred to as intelligent ultrasound system) in adult goiter.Methods:In June 2022 and March 2023, two phases of thyroid disease survey were carried out in 4 cities in Anhui Province. One village was selected in each city, and 250 adults were selected as survey subjects in each village. Adult bilateral thyroid area was scanned by both intelligent ultrasound system and conventional ultrasound scanning equipment, and the effectiveness of intelligent ultrasound system in the diagnosis of goiter was analyzed based on the results of conventional ultrasound examination. Receiver operating characteristic (ROC) curve was drawn, and Kappa test was used to analyze the consistency between intelligent ultrasound system and conventional ultrasound examination in the diagnosis of goiter. At the same time, Spearman correlation analysis and Bland-Altman method were used to evaluate the consistency of the two methods in measuring thyroid volume.Results:After screening and removing outliers and missing values, a total of 910 adults were included, including 253 males (27.80%) and 657 females (72.20%). The age was (45.92 ± 10.20) years old, ranging from 18 to 60 years old. The sensitivity, specificity, and accuracy of the intelligent ultrasound system for diagnosing adult goiter were 80.00%, 99.67%, and 99.56%, respectively. The area under the ROC curve (AUC) was 0.996, which was consistent with the results of conventional ultrasound examination for diagnosing goiter ( κ = 0.67, P < 0.001). After controlling for variables such as gender, thyroid function, and thyroid nodules, the intelligent ultrasound system showed good consistency with conventional ultrasound examination in the diagnosis of goiter in females, adults with thyroid dysfunction, and adults without thyroid nodules ( κ = 0.66, 0.80, 0.80, P < 0.001). The consistency in the diagnosis of goiter in adults with thyroid nodules was moderate ( κ = 0.56, P < 0.001). Spearman correlation analysis showed a highly positive correlation between the measurement results of adult thyroid volume by intelligent ultrasound system and conventional ultrasound examination ( r = 0.88, P < 0.001). The Bland-Altman method results showed that only 4.62% (42/910) of points in adults were outside the 95% consistency limit, indicating good consistency between intelligent ultrasound system and conventional ultrasound examination in measuring thyroid volume (< 5%). The proportion of points outside the 95% consistency limit in males, adults with thyroid dysfunction, and adults with thyroid nodules was 6.72% (17/253), 5.83% (12/206), and 6.45% (12/186), respectively. Conclusions:The intelligent ultrasound system has certain diagnostic value for adult goiter and has good consistency with conventional ultrasound examination for thyroid volume measurement. However, the accuracy of diagnosis for males and adults with thyroid nodules still needs to be improved.

ObjectiveTo reveal the targets and molecular mechanisms of the action of Qiangxin Decoction （强心汤） for the treatment of chronic heart failure based on the combination of network pharmacology and molecular docking. MethodsThe active ingredients of Qiangxin Decoction were retrieved from TCMSP database， and the targets of chronic heart failure were screened by searching GeneCards， OMIM， TTD， PharmGkb， and DrugBank databases， and the intersections were taken to obtain the intersecting targets of Qiangxin Decoction for the treatment of chronic heart failure. STRING platform was used to construct the protein-protein interaction network （PPI）， Cytoscape 3.8.0 software was used to calculate the network topology to screen the core targets， and R 4.2.3 was used to construct the “active ingredient-target” network by analyzing the GO enrichment analysis and KEGG pathway enrichment analysis. AutoDock 1.5.7 was used for molecular docking to predict the binding performance of active ingredients and core targets. ResultsSeventy-five intersecting targets were identified for the treatment of chronic heart failure with Qiangxin Decoction， among which the core targets were estrogen receptor 1 （ESR1， degree value=7）， nuclear receptor coactivator 1 （NCOA1， degree value=8）， glucocorticoid receptor （NR3C1， degree value=7）， and nuclear receptor coactivator 2 （NCOA2， degree value=7）. GO enrichment analysis showed that the top 3 items with the smallest P value in molecular function were G protein-coupled amine receptor activity， postsynaptic neurotransmitter receptor activity， and neurotransmitter receptor activity （P＜0.01）； the top 3 items with the smallest P value in biological process were adenylyl cyclase-activated adrenergic receptor signaling pathway， adrenergic receptor signaling pathway， and adenylyl cyclase-regulated G protein-coupled receptor signaling pathway （P＜0.01）； the top 3 items with the smallest P values in cellular composition were components of the postsynaptic membrane， synaptic membrane， and presynaptic membrane （P＜0.01）. KEGG enrichment analysis showed that the top 5 key signaling pathways were neuroactive ligand-receptor interactions， calcium signaling pathway， dopaminergic synapses， cocaine addiction， and cyclic guanosine monophosphate-protein kinase G （cGMP-PKG） signaling pathway. The molecular docking results showed that lignans and isoflavones had lower binding energies and more structural stability with the four core targets （ESR1， NCOA1， NR3C1， NCOA2）. ConclusionThe treatment of chronic heart failure by Qiangxin Decoction was associated with neuroactive ligand-receptor interactions， calcium signaling pathway， dopaminergic synapses， chemoattractant-receptor activation， cGMP-PKG signaling pathway， lipids and atherosclerosis， and cAMP signaling pathway， and lignans and isoflavones may be the core active compounds in its treatment of chronic heart failure.

Objective:Multidrug-resistant tuberculosis(MDR-TB)has a high mortality and is always one of the major challenges in global TB prevention and control.Analyzing the factors that may impact the adverse outcomes of MDR-TB patients is helpful for improving the systematic management and optimizing the treatment strategies for MDR-TB patients.For follow-up data,the Cox proportional hazards regression model is an important multifactor analysis method.However,the method has significant limitations in its application,such as the fact that it is difficult to deal with the impacts of small sample sizes and other practical issues on the model.Therefore,Bayesian and conventional Cox regression models were both used in this study to analyze the influencing factors of death in MDR-TB patients during the anti-TB therapy,and compare the differences between these 2 methods in their application. Methods:Data were obtained from 388 MDR-TB patients treated at Lanzhou Pulmonary Hospital from November 1,2017 to March 31,2021.Survival analysis was employed to analyze the death of MDR-TB patients during the therapy and its influencing factors.Conventional and Bayesian Cox regression models were established to estimate the hazard ratios(HR)and their 95% confidence interval(95% CI)for the factors affecting the death of MDR-TB patients.The reliability of parameter estimation in these 2 models was assessed by comparing the parameter standard deviation and 95% CI of each variable.The smaller parameter standard deviation and narrower 95% CI range indicated the more reliable parameter estimation. Results:The median survival time(1st quartile,3rd quartile)of the 388 MDR-TB patients included in the study was 10.18(4.26,18.13)months,with the longest survival time of 31.90 months.Among these patients,a total of 12 individuals died of MDR-TB and the mortality was 3.1%.The median survival time(1st quartile,3rd quartile)for the deceased patients was 4.78(2.63,6.93)months.The majority of deceased patients,accounting for 50%,experienced death within the first 5 months of anti-TB therapy,with the last mortality case occurring within the 13th month of therapy.The results of the conventional Cox regression model showed that the risk of death in MDR-TB patients with comorbidities was approximately 6.96 times higher than that of patients without complications(HR=6.96,95% CI 2.00 to 24.24,P=0.002)and patients who received regular follow-up had a decrease in the risk of death by approximately 81% compared to those who did not receive regular follow-up(HR=0.19,95% CI 0.05 to 0.77,P=0.020).In the results of Bayesian Cox regression model,the iterative history plot and Blue/Green/Red(BGR)plot for each parameter showed the good model convergence,and parameter estimation indicated that the risk of death in patients with a positive first sputum culture was lower than that of patients with a negative first sputum culture(HR=0.33,95% CI 0.08 to 0.87).Additionally,compared to patients without complications,those with comorbidities had an approximately 6.80-fold increase in the risk of death(HR=7.80,95% CI 1.90 to 21.91).Patients who received regular follow-up had a 90% reduction in the risk of death compared to those who did not receive regular follow-up(HR=0.10,95% CI 0.01 to 0.30).The comparison between these 2 models showed that the parameter standard deviations and corresponding 95% CI ranges of other variables in the Bayesian Cox model were significantly smaller than those in the conventional model,except for parameter standard deviations of receiving regular follow-up(Bayesian model was 0.77;conventional model was 0.72)and pulmonary cavities(Bayesian model was 0.73;conventional model was 0.73). Conclusion:The first year of anti-TB therapy is a high-risk period for mortality in MDR-TB patients.Complications are the main risk factors of death in MDR-TB patients,while patients who received regular follow-up and had positive first sputum culture presented a lower risk of death.For data with a small sample size and low incidence of outcome,the Bayesian Cox regression model provides more reliable parameter estimation than the conventional Cox model.

Objective:To explore the value of radiomics model based on intratumoral and peritumoral early dynamic contrast-enhanced (DCE) MRI for identifying benign and malignant in breast imaging reporting and data system (BI-RADS) 4 tumors.Methods:A total of 191 patients diagnosed with BI-RADS 4 breast tumors by breast MRI examination with clear pathological diagnosis from January 2016 to December 2020 in the First Affiliated Hospital of Bengbu Medical College were analyzed retrospectively, including 77 benign and 114 malignant cases, aged 23-68 (46±10) years. The one-slice image with the largest area of the lesion of the second stage DCE-MRI images was selected to outline the region of interest, and automatically conformal extrapolated by 5 mm to extract the intra-tumoral and peritumoral radiomics features. The included cases were randomly divided into training and testing cohorts in the ratio of 8∶2. The statistical and machine learning methods were used for feature dimensionality reduction and selection of optimal radiomics features, and logistic regression was used as the classifier to establish the intratumoral, peritumoral, and intratumoral combined with peritumoral radiomics models. The independent risk factors that could predict the benignity and malignancy of breast tumors were retained as clinical-radiological characteristics by univariate and multivariate logistic regression to establish a clinical-radiological model. Finally, the intratumoral and peritumoral radiomics features were combined with clinical-radiological features to develop a combined model of the three. The receiver operating curve was used to analyze the predictive performance of each model and calculate the area under the curve (AUC),the AUC was compared by DeLong test. The stability of the three-component combined diagnostic model was tested by 10-fold cross-validation, and the model was visualized by plotting nomogram and calibration curves.Results:In the training cohort, the AUC of the three-component combined model for identifying benign and malignant BI-RADS 4 breast tumors was significantly higher than that of the intratumoral radiomics model ( Z=3.38, P<0.001), the peritumoral radiomics model ( Z=4.01, P<0.001), the intratumoral combined with peritumoral radiomics model ( Z=3.11, P=0.002), and the clinical-radiological model ( Z=3.24, P=0.001). And the AUC, sensitivity, specificity, accuracy, and F1-score of the three-component combined model were 0.932, 91.2%, 86.9%, 87.0% and 0.89, respectively. In the testing cohort, the three-component combined model also had the highest AUC value (0.875), and diagnostic sensitivity, specificity, accuracy and malignancy F1-score were 95.7%, 62.5%, 76.9%, and 0.89, respectively. The AUC calculated by 10-fold cross-validation was 0.90 (0.85-0.92), and the predicted curve of the three-component combined model in the calibration curve was in good agreement with the ideal curve. Conclusion:The three-component combined diagnostic model based on the intratumoral and peritumoral radiomics features and clinical-radiological features of early DCE-MRI has good performance and stability for identifying the benign and malignant in BI-RADS 4 breast tumors, and it can provide guidance for clinical decision non-invasively.

Objective:To compare the clinical characteristics of critically ill pregnant women admitted to the intensive care unit (ICU) with different admission methods, in order to make more effective and rational use of ICU resources.Methods:A retrospective study was conducted. The clinical data of critically ill pregnant women admitted to ICU of Peking University Third Hospital from January 2006 to July 2019 were analyzed. According to the admission mode to ICU, the pregnant women were divided into emergency admission group (transferred to ICU on the same day or the next day due to critical illness) and planned admission group (transferred to ICU 2 days after admitting in obstetric ward). The clinical characteristics of ICU critical pregnant women, such as the incidence, causes of admission, severity of the disease, main treatment measures, mortality, and medical expenses were collected, and a comparative analysis between the two groups was performed.Results:During the nearly 14 years, a total of 576 critical pregnant women in ICU were enrolled, accounting for 0.8% (576/71 790) of the total number of obstetric inpatients and 4.6% (576/12 412) of the total number of ICU inpatients. Seven maternal deaths accounted for 1.2% of all critically pregnant women transferred to ICU, and the overall mortality of pregnant women was 10/100 thousand. Of the 576 critically pregnant women, there were 327 patients (56.8%) in the emergency admission group and 249 patients (43.2%) in the planned admission group. Compared with the planned admission group, the proportion of elective cesarean section in the emergency admission group was significantly lower (17.7% vs. 94.0%, P < 0.01), and the proportion of emergency cesarean section was significantly higher (65.1% vs. 2.4%, P < 0.01), the acute physiology and chronic health evaluation (APACHE Ⅱ, APACHE Ⅲ) scores, simplified acute physiology score Ⅱ (SAPS Ⅱ) and Marshall score were significantly higher [APACHE Ⅱ score: 6.0 (4.0, 9.8) vs. 4.0 (3.0, 7.0), APACHE Ⅲ score: 14.0 (11.0, 20.3) vs. 12.0 (9.0, 16.0), SAPS Ⅱ score: 8 (0, 12) vs. 3 (0, 8), Marshall score: 2 (1, 4) vs. 1 (1, 3), all P < 0.01]. The length of ICU stay in the emergency admission group was significantly longer than that in the planned admission group [days: 2 (1, 5) vs. 2 (1, 3), P < 0.01], and the total length of hospital stay was significantly shorter [days: 9 (7, 13) vs. 13 (10, 18), P < 0.01]. Both in the emergency admission group and the planned admission group, obstetric factors were the main reason for admission, 60.9% (199/327) and 70.3% (175/249), respectively. The proportion of postpartum hemorrhage was the highest [35.2% (115/327) and 57.0% (142/249)], followed by preeclampsia/eclampsia [7.0% (23/327) and 7.6% (19/249)]. Only 7 of the 19 critically pregnant women with puerperal infection were planned admission. All 21 patients with acute fatty liver of pregnancy (AFLP) during pregnancy were emergency admission. Among the emergency and planned admission patients, 73 patients (22.3%) and 42 patients (16.9%) required mechanical ventilation (duration of mechanical ventilation > 24 hours), 99 patients (30.3%) and 35 patients (14.1%) needed vasoactive agents, 67 patients (20.5%) and 20 patients (8.0%) received hemodynamic monitoring, and 123 patients (37.6%) and 154 patients (61.8%) were given anticoagulation therapy, respectively. In terms of severity score of critical pregnant women, there were significant differences in APACHE Ⅱ, APACHE Ⅲ, SAPS Ⅱ and Marshall scores of pregnant women with different diseases. Among them, the APACHE Ⅲ, SAPS Ⅱ and Marshall scores of AFLP were the highest [21.0 (15.0, 32.5), 12.0 (6.0, 16.5) and 6.0 (3.5, 8.0), respectively]. The APACHE Ⅱ and APACHE Ⅲ scores of postpartum hemorrhage were the lowest [4.0 (3.0, 7.0), 12.0 (10.0, 16.0)]. The SAPS Ⅱ score of pneumonia was the lowest [2.0 (0, 14.0)]. The Marshall score for puerperal infection was the lowest [1.0 (0, 3.0)]. In terms of the total medical expenses, the cost in the emergency admission group was significantly lower than that in the planned admission group [10 thousand Yuan: 3.1 (2.0, 4.7) vs. 4.1 (2.9, 5.8), P < 0.05]. Conclusions:Compared with the critically ill pregnant women who planned to be admitted to ICU, the patients emergency admitted to ICU were more complicated and urgent, and the severity of the condition was scored higher. At present, the severity scoring system commonly used in ICU can only partly evaluate the severity of critically ill pregnant women, therefore, it is necessary to design the specific severity scoring system for critically ill pregnant women to effectively and rationally use the precious ICU resources.