Objective:To fine-tune the mT5 (massively multilingual pre-trained text-to-text transformer) large language model, automatically generate report conclusions for teaching purposes from chest CT image descriptions, and assess the quality of automatically generated conclusions.Methods:The training set included 3 000 high-quality physical examination chest CT reports from one hospital, and the external validation set consisted of 600 physical examination chest CT reports from two other hospitals. Experienced radiology teaching physicians assessed the consistency between the generated conclusions and the original physician-written conclusions in the external validation set using a 5-point Likert scale across five linguistic indicators (correctness of examination information, correctness of lesion detection, standardization of terminology, applicability of the conclusions, and simplicity of conclusions). Using the original report conclusions as the reference, the accuracy of the conclusions generated based on the external validation set in describing four major thoracic conditions (pulmonary nodules, pneumonia, emphysema, pleural effusion) was evaluated. Perform chi square test using SPSS 25.0.Results:In the external validation set, the mean consistency score between the generated conclusions and the original conclusions given by the radiology teaching physicians was >4 points, indicating agreement with the original conclusions. In the generated conclusions, the description of the four major thoracic conditions demonstrated 0.95-1.00 (95% CI=0.91-1.00) accuracy, 0.76-1.00 (95% CI=0.59-1.00) sensitivity, and 0.97-1.00 (95% CI=0.91-1.00) specificity. Conclusions:The chest CT report conclusion generation system based on the mT5 large language model demonstrated high accuracy and is expected to provide immediate and efficient automated guidance for standardized residency training.

Protein liquid-liquid phase separation (LLPS), a pivotal phenomenon intricately linked to cellular processes, is regulated by various other proteins. However, there is still a lack of high-throughput methods for screening protein regulators of LLPS in target proteins. Here, we developed a CRISPR/Cas9-based screening method to identify protein phase separation regulators by integrating bimolecular fluorescence complementation (BiFC) and fluorescence-activated cell sorting (FACS). Using this newly developed method, we screened the RNA-binding proteins that regulate PABPN1 phase separation and identified the tumor suppressor QKI as a promoter of PABPN1 phase separation. Furthermore, QKI exhibits decreased expression levels and diminished nuclear localization in colorectal cancer cells, resulting in reduced PABPN1 phase separation, which, in turn, promotes alternative polyadenylation (APA), cell proliferation, and migration in colorectal cancer.
Humans ; Colorectal Neoplasms/genetics* ; RNA-Binding Proteins/genetics* ; Poly(A)-Binding Protein I/genetics* ; CRISPR-Cas Systems ; Flow Cytometry ; Cell Proliferation ; Cell Line, Tumor ; Cell Movement

To investigate the awareness of cognitive impairment disorders among residents of the Xizang Autonomous Region and its influencing factors, thereby providing a basis for targeted prevention and treatment efforts.

The machine learning is used increasingly and widely in acupuncture prescription optimization, intelligent treatment and precision medicine, and has obtained a certain achievement. But, there are still some problems remained to be solved such as the poor interpretability of the model, the inconsistency of data quality of acupuncture research, and the clinical application of constructed models. Researches in future should focus on the acquisition of high-quality clinical and experimental data sets, take various machine learning algorithms as the basis, and construct professional models to solve various problems, so as to drive the high-quality development of acupuncture research.
Acupuncture Therapy/trends* ; Machine Learning ; Humans ; Algorithms

Objective To understand the current status of illness benefit perception among caregivers of type 2 diabetes patients,analyze its relationship with coping strategies and social support,and explore its influencing factors.Methods Using a convenience sampling method,240 caregivers of type 2 diabetes patients hospitalized in the endocrinology department of a tertiary hospital in Kunming from October 2023 to October 2024 were selected.Questionnaires including general information survey,Benefit Finding Scale,Simplified Coping Strategy Scale,and Social Support Rating Scale were used to investigate the caregivers.Results The caregivers'benefit finding score was(66.82±16.73),coping strategy score was(36.01±9.64),and social support score was(42.00±8.11).Pearson correlation analysis showed positive correlations between coping styles,social support,and illness benefit perception(r=0.445,r=0.493,P<0.01);multiple linear regression analysis indicated that family residence,education level,duration of caregiving,and understanding of the patient's disease were key factors influencing illness benefit perception(P<0.01).Conclusion Caregivers of type 2 diabetes patients have a moderate level of illness benefit perception,with positive correlations between coping strategies,social support,and illness benefit perception.The disease benefit perception of caregivers is also influenced by family residence,education level,caregiving duration,and disease knowledge.

Objective To explore the application value of artificial intelligence in medical research assistance, and analyze the key paths to achieve precise execution of model instructions, improvement of model interpretation completeness, and control of hallucinations. Methods Taking esophageal cancer research as the scenario, five types of literature including research articles, case reports, reviews, editorials, and guidelines were selected for model interpretation tests. The model performance was systematically evaluated from five dimensions: recognition accuracy, format accuracy, instruction execution accuracy, content reliability rate, and content completeness index. The performance differences of Ruibin Agent, GPT-4o, Claude 3.7 Sonnet, DeepSeek V3, and DouBao-pro models in medical literature interpretation tasks were compared. Results A total of 15 studies were included, with 3 studies of each type. The five models collectively conducted 1 875 tests. Due to the poor recognition accuracy of the editorial type, the overall recognition accuracy of Ruibin Agent was significantly lower than other models (92.0% vs. 100.0%, P＜0.001). In terms of format accuracy, Ruibin Agent was significantly better than Claude 3.7 Sonnet (98.7% vs. 92.0%, P=0.002) and GPT-4o (98.7% vs. 78.9%, P＜0.001). In terms of instruction execution accuracy, Ruibin Agent was better than GPT-4o (97.3% vs. 80.0%, P＜0.001). In terms of content reliability rate, Ruibin Agent was significantly lower than Claude 3.7 Sonnet (84.0% vs. 92.0%, P=0.010) and DeepSeek V3 (84.0% vs. 94.7%, P＜0.001). In terms of content completeness index, the median scores of Ruibin Agent, GPT-4o, Claude 3.7 Sonnet, DeepSeek V3, and DouBao-pro were 0.71, 0.60, 0.85, 0.74, and 0.77, respectively. Conclusion Ruibin Agent has significant advantages in terms of formatted interpretation of medical literature and instruction execution accuracy. In the future, it is necessary to focus on optimizing the recognition ability of editorial types, strengthening the coverage ability of core elements of various types of literature to improve interpretation completeness, and improving content reliability through optimizing the confidence mechanism to ensure the rigor of medical literature interpretation.

Objective To investigate the achievement of early fluid resuscitation targets and factors influencing 28-day outcomes in patients with sepsis.Methods A retrospective cohort analysis was conducted.A total of 164 patients with sepsis admitted to the First Affiliated Hospital of Xinjiang Medical University between January 2022 and January 2024 were enrolled.Patients were divided into survival and death groups based on 28-day survival status,with both groups receiving early fluid resuscitation.Comparisons were made between groups for general characteristics[gender,age,body mass index(BMI),infection site,comorbidities],primary indicators[central venous pressure(CVP),mean arterial pressure(MAP),urine output],and secondary indicators[blood lactate acid(Lac),procalcitonin(PCT),heart rate,sequential organ failure assessment(SOFA)on intensive care unit(ICU)admission day,Glasgow coma scale(GCS),duration and dose of vasoactive medication use].Univariate analysis identified variables associated with prognosis,followed by multivariate Logistic regression to select independent risk factors.Receiver operator characteristic(ROC curve)were plotted to assess predictive performance of each risk factor for the 28-day prognosis of patients with sepsis.Results This study included 164 patients.The primary infection sites were mainly the lungs,abdominal cavity,and urinary system,accounting for 42.7%(70/164),38.4%(63/164),and 9.1%(15/164)respectively.The survival group comprised 141 patients,while the the death group included 23 patients.No statistically significant differences existed between groups in gender,BMI,infection site(soft tissue infection vs.others),underlying diseases,MAP,urine output(all P>0.05).Compared to the survival group,the death group showed significantly higher age,pulmonary infection rate,Lac levels,vasoactive drug duration/dose,heart rate,and SOFA scores,while the rates of abdominal,and urinary tract infection,as well as CVP,PCT,and GCS scores were significantly lower(all P<0.05).The achievement rates of early fluid resuscitation parameters:MAP target achievement was highest at 78.7%(129/164),followed by urine output compliance at 78.0%(128/164),while CVP compliance was the lowest at 39.0%(64/164).The overall compliance rate was 21.3%(35/164).Univariate analysis showed that age,pulmonary infection,Lac levels,duration and dose of vasoactive drugs,heart rate,PCT,GCS score,and SOFA score were all risk factors affecting the 28-day prognosis of patients with sepsis(all P<0.05).Multivariate Logistic regression analysis showed Lac levels,and pulmonary infection were independent risk factors affecting 28-day prognosis of patients with sepsis[odds ratio(OR)were 0.801,3.966,0.812,95%confidence interval(95%CI)were 0.711-0.903,1.149-13.696,0.674-0.979 respectively,P values were<0.001,0.029,0.029 respectively].ROC curve analysis demonstrated that age,Lac levels,and pulmonary infection all possessed predictive value for 28-day outcomes(all P<0.05).Age exhibited the highest predictive value with an AUC of 0.922.At the optimal cut-off of 76.6 years,sensitivity reached 95.7%and specificity 80.9%.Conclusion The overall achievement rate of early fluid resuscation in sepsis patients was low,with age,Lac levels,and pulmonary infection being major factors influencing poor prognosis.

Objective:To fine-tune the mT5 (massively multilingual pre-trained text-to-text transformer) large language model, automatically generate report conclusions for teaching purposes from chest CT image descriptions, and assess the quality of automatically generated conclusions.Methods:The training set included 3 000 high-quality physical examination chest CT reports from one hospital, and the external validation set consisted of 600 physical examination chest CT reports from two other hospitals. Experienced radiology teaching physicians assessed the consistency between the generated conclusions and the original physician-written conclusions in the external validation set using a 5-point Likert scale across five linguistic indicators (correctness of examination information, correctness of lesion detection, standardization of terminology, applicability of the conclusions, and simplicity of conclusions). Using the original report conclusions as the reference, the accuracy of the conclusions generated based on the external validation set in describing four major thoracic conditions (pulmonary nodules, pneumonia, emphysema, pleural effusion) was evaluated. Perform chi square test using SPSS 25.0.Results:In the external validation set, the mean consistency score between the generated conclusions and the original conclusions given by the radiology teaching physicians was >4 points, indicating agreement with the original conclusions. In the generated conclusions, the description of the four major thoracic conditions demonstrated 0.95-1.00 (95% CI=0.91-1.00) accuracy, 0.76-1.00 (95% CI=0.59-1.00) sensitivity, and 0.97-1.00 (95% CI=0.91-1.00) specificity. Conclusions:The chest CT report conclusion generation system based on the mT5 large language model demonstrated high accuracy and is expected to provide immediate and efficient automated guidance for standardized residency training.

Objective To investigate the achievement of early fluid resuscitation targets and factors influencing 28-day outcomes in patients with sepsis.Methods A retrospective cohort analysis was conducted.A total of 164 patients with sepsis admitted to the First Affiliated Hospital of Xinjiang Medical University between January 2022 and January 2024 were enrolled.Patients were divided into survival and death groups based on 28-day survival status,with both groups receiving early fluid resuscitation.Comparisons were made between groups for general characteristics[gender,age,body mass index(BMI),infection site,comorbidities],primary indicators[central venous pressure(CVP),mean arterial pressure(MAP),urine output],and secondary indicators[blood lactate acid(Lac),procalcitonin(PCT),heart rate,sequential organ failure assessment(SOFA)on intensive care unit(ICU)admission day,Glasgow coma scale(GCS),duration and dose of vasoactive medication use].Univariate analysis identified variables associated with prognosis,followed by multivariate Logistic regression to select independent risk factors.Receiver operator characteristic(ROC curve)were plotted to assess predictive performance of each risk factor for the 28-day prognosis of patients with sepsis.Results This study included 164 patients.The primary infection sites were mainly the lungs,abdominal cavity,and urinary system,accounting for 42.7%(70/164),38.4%(63/164),and 9.1%(15/164)respectively.The survival group comprised 141 patients,while the the death group included 23 patients.No statistically significant differences existed between groups in gender,BMI,infection site(soft tissue infection vs.others),underlying diseases,MAP,urine output(all P>0.05).Compared to the survival group,the death group showed significantly higher age,pulmonary infection rate,Lac levels,vasoactive drug duration/dose,heart rate,and SOFA scores,while the rates of abdominal,and urinary tract infection,as well as CVP,PCT,and GCS scores were significantly lower(all P<0.05).The achievement rates of early fluid resuscitation parameters:MAP target achievement was highest at 78.7%(129/164),followed by urine output compliance at 78.0%(128/164),while CVP compliance was the lowest at 39.0%(64/164).The overall compliance rate was 21.3%(35/164).Univariate analysis showed that age,pulmonary infection,Lac levels,duration and dose of vasoactive drugs,heart rate,PCT,GCS score,and SOFA score were all risk factors affecting the 28-day prognosis of patients with sepsis(all P<0.05).Multivariate Logistic regression analysis showed Lac levels,and pulmonary infection were independent risk factors affecting 28-day prognosis of patients with sepsis[odds ratio(OR)were 0.801,3.966,0.812,95%confidence interval(95%CI)were 0.711-0.903,1.149-13.696,0.674-0.979 respectively,P values were<0.001,0.029,0.029 respectively].ROC curve analysis demonstrated that age,Lac levels,and pulmonary infection all possessed predictive value for 28-day outcomes(all P<0.05).Age exhibited the highest predictive value with an AUC of 0.922.At the optimal cut-off of 76.6 years,sensitivity reached 95.7%and specificity 80.9%.Conclusion The overall achievement rate of early fluid resuscation in sepsis patients was low,with age,Lac levels,and pulmonary infection being major factors influencing poor prognosis.

Objective:To investigate the clinical features of nontuberculous mycobacteria (NTM) disease patients with positive anti-interferon γ (IFN-γ) autoantibody.Methods:Forty-three adult human immunodeficiency virus-uninfected patients with NTM disease hospitalized in Huashan Hospital, Fudan University and Jing′an Branch, Huashan Hospital, Fudan University from July 2021 to August 2023 were included. Clinical data and NTM strain information of the patients were collected. The plasma levels of anti-IFN-γ autoantibodies were detected by enzyme-linked immunosorbent assay, and the patients were divided into antibody positive group and antibody negative group. The clinical characteristics and laboratory examination results between the two groups were compared. The independent sample t test, Mann-Whitney U test and chi-square test were used for statistical analysis. Multivariate logistic regression analysis was used to determine the correlation factors of positive anti-IFN-γ autoantibodies. Results:Among the 43 patients, 13 cases (30.2%) were positive for anti-IFN-γ autoantibodies and 30 cases (69.8%) were negative. The proportions of patients with NTM disseminated infection (9/13 vs 30.0%(9/30))and combined bacterial infection (5/13 vs 6.7%(2/30)) in antibody positive group were both higher than those in antibody negative group, and the differences were both statistically significant ( χ2=5.74 and 6.73, respectively, both P<0.05). The white blood cell count, platelet count, the proportion of platelet count >350×10 9/L of antibody positive patients were all higher than those of antibody negative group, while the white sphere ratio was lower than that of antibody negative group, with statistical significance ( t=2.42, 3.02, χ2=9.77 and t=3.66, respectively, all P<0.05). Erythrocyte sedimentation rate, C-reactive protein, procalcitonin, globulin, immunoglobulin G, immunoglobulin A and immunoglobulin M in antibody positive patients were all higher than those in antibody negative group, and the differences were all statistically significant ( U=99.50, 112.00, 115.50, 61.50, 76.50, 99.00 and 83.00, respectively, all P<0.05). Mycobacterium abscessus complex (seven cases and 11 cases, respectively) and Mycobacterium avium complex (five cases and 13 cases, respectively) were the main isolated strains in antibody positive and antibody negative patients. Multivariate logistic regression analysis showed that combined with bacterial infection (odds ratio ( OR)=21.83, 95% confidence interval ( CI) 1.94 to 245.71), NTM disseminated infection ( OR=7.64, 95% CI 1.10 to 53.26), platelet count>350×10 9/L ( OR=14.31, 95% CI 1.91 to 107.04) were risk factors for anti-IFN-γ autoantibodies positive (all P<0.05). Conclusions:Patients with positive anti-IFN-γ autoantibodies have higher probability of having elevated levels of systemic inflammation. Anti-IFN-γ autoantibody test is recommended for patients with NTM disease who present with co-bacterial infection, NTM disseminated infection, or elevated platelet count (>350×10 9/L).