OBJECTIVES: Accurate identification of risk factors for precocious puberty is essential for clinical diagnosis and management, yet the performance of natural language processing methods applied to unstructured electronic medical record (EMR) data remains to be fully evaluated. This study aims to assess the performance of a prompt engineering method for extracting individual risk factors of precocious puberty from EMRs. METHODS: Based on the capacity and role-insight-statement-personality-experiment (CRISPE) prompt framework, both simple and optimized prompts were designed to guide the large language model GLM-4-9B in extracting 10 types of risk factors for precocious puberty from 653 EMRs. Accuracy, precision, recall, and F1-score were used as evaluation metrics for the information extraction task. RESULTS: Under simple and optimized prompt conditions, the overall accuracy, precision, recall, and F1-score of the model were 84.18%, 98.09%, 81.99%, and 89.32% versus 97.15%, 98.31%, 98.16%, and 98.23%, respectively. The optimized prompts achieved more stable performance across age (<9 years vs ≥9 years) and visit-time (<2023 vs ≥2023) subgroups compared with simple prompts. The accuracy range for extracting each risk factor was 60.03%-97.24%, while with optimized prompts, the range improved to 92.19%-99.85%. The largest performance improvement occurred for "beverage intake" (60.03% vs 92.19%), and the smallest for "maternal age of menarche" (97.24% vs 99.23%). In comparing distributions among simple prompts, optimized prompts, and ground truth, statistically significant differences were observed for snack intake, beverage intake, soy milk intake, honey intake, supplement use, tonic use, sleep quality, and sleeping with the light on (all P<0.001), while exercise (P=0.966) and maternal menarche age (P=0.952) showed no significant differences. CONCLUSIONS Compared with simple prompts, optimized prompts substantially improved the extraction performance of individual risk factors for precocious puberty from EMRs, underscoring the critical role of prompt engineering in enhancing large language model performance.
Humans ; Puberty, Precocious/epidemiology* ; Risk Factors ; Electronic Health Records ; Female ; Child ; Natural Language Processing

Human NAD(P)H: quinone oxidoreductase 1 (NQO1) is a flavoenzyme expressed at high levels in multiple solid tumors, making it an attractive target for anticancer drugs. Bioactivatable drugs targeting NQO1, such as β-lapachone (β-lap), are currently in clinical trials for the treatment of cancer. β-Lap selectively kills NQO1-positive (NQO1⁺) cancer cells by inducing reactive oxygen species (ROS) via catalytic activation of NQO1. In this study, we demonstrated that cryptotanshinone (CTS), a naturally occurring compound, induces NQO1-dependent necrosis without affecting NQO1 activity. CTS selectively kills NQO1⁺ cancer cells by inducing NQO1-dependent necrosis. Interestingly, CTS directly binds to NQO1 but does not activate its catalytic activity. In addition, CTS enables activation of JNK1/2 and PARP, accumulation of iron and Ca²⁺, and depletion of ATP and NAD⁺. Furthermore, CTS selectively suppressed tumor growth in the NQO1⁺ xenograft models, which was reversed by NQO1 inhibitor and NQO1 shRNA. In conclusion, CTS induces NQO1-dependent necrosis via the JNK1/2/iron/PARP/NAD⁺/Ca²⁺ signaling pathway. This study demonstrates the non-enzymatic function of NQO1 in inducing cell death and provides new avenues for the design and development of NQO1-targeted anticancer drugs.

Image 1.

Objective: To investigate the incidence and influencing factors of falls among the elderly in Ningbo City, Zhejiang Province, so as to provide the basis for developing effective prevention strategies. Methods: The residents aged 60 years and above in Haishu District and Yuyao City of Ningbo City were selected by the multi-stage cluster random sampling method from June to October 2022. Demographic information, fall incidence in the past year, history of disease and self-rated health were collected through questionnaire surveys. Incidence of falls was descriptively analyzed, and factors affecting falls were identified using a multivariable logistic regression model stratified by gender and age. Results: A total of 1 275 elderly people were surveyed, including 635 men and 640 women. The median age was 72.00 (interquartile range, 13.00) years. In the past year, 158 residents fell, accounting for 12.39%. Additionally, 14 individuals experienced two or more falls, accounting for 8.86%. The incidence of falls was 14.69% in women, which was higher than the 10.08% in men (P<0.05). The incidence of falls was 14.86% in the elderly over 70 years, which was higher than the 9.39% in those aged 60 to 70 years (P<0.05). Multivariable logistic regression showed that the educational level (primary school and above, OR=0.501, 95%CI: 0.301-0.836), heart disease (present, OR=1.996, 95%CI: 1.076-3.703), and self-rated health status (good, OR=0.529, 95%CI: 0.319-0.875) were factors affecting falls in women; educational level (primary school and above, OR=0.514, 95%CI: 0.285-0.928) and self-rated health status (good, OR=0.456, 95%CI: 0.253-0.824) were factors affecting falls in residents aged 60 to 70 years. Conclusion Fall risk among the elderly is associated with gender, age, heart disease, educational level and self-rated health status, and the influencing factors for falls vary in different genders and ages.

Objective To analyze the clinical characteristics of IgG4-related disease (IgG4-RD),guide the selection of therapeutic drugs,and to explore the significance of potential tumor identification for IgG4-RD.Methods A total of 92 patients diagnosed with IgG4-RD and admitted to this hospital from January 1,2017 to December 31,2021were selected as the research subjects by using the Yidu Cloud system.The clinical data conducted the summary analysis. The clinical characteristics of IgG4-RD were summarized.Results The mean age of IgG4-RD was definitely diagnosed in the 92 patients was (58.1±11.3)years old,with 65 male ca-ses (70.7％) and 27 female cases (29.3％).The most commonly affected organ tissues were lymph nodes (37 cases,40.2％),pancreas (33 cases,35.9％) and salivary glands (31 cases,33.7％).In the patients woth the 92 patients,28 cases (30.4％) had involvement of a single organ tissue,while 32 cases (34.8％) had involvement of two or more organs.In the 92 patients,89 cases received steroid therapy,and 71 cases received immunosup-pressive therapy,in which 45 cases (63.4％) used cyclophosphamide.The initial treatment effective rate (72.7％ vs. 55.6％) and one-year non-recurrence rate (38.2％ vs. 20.0％) of the steroid combined immuno-suppressive therapy group were better than those of the single steroid group,but the differences were not sta-tistically significant (P>0.05).The proportion of the patients with tumor comorbidity and IgG4 level>40 g/L (18.2％) was significantly higher than that of the non-tumor comorbidity (1.2％),and the difference was statistically significant (P<0.05).However,there was no statistically significant difference in the proportion of patients with tumor comorbidity compared to the non-tumor comorbidity in other IgG4 level groups (P>0.05).Conclusion IgG4-RD is more common in middle-aged and elderly men,lymph nodes,pancreas and sal-ivary glands are commonly involved,and most patients have the double organs and multiple organs involve-ment. The combination use of hormone and immunosuppressant in treatment is recommended .The IgG4 lev-el>40 g/L in the patients with IgG4-RD may has the suggestive significance for complicating tumor.

0n December 31,2021,the Chinese Nursing Association released the group standard"nursing care of patients with infusion of vasoactive agents(T/CNAS 22-2021)",which outlines the fundamental requirements for intravenous infusion of vasoactive drugs and standardizes the evaluation,administration,and monitoring.This article provides an interpretation of the key parts and sections of the standard to ensure nursing safety during the administration of vasoactive drugs,aiming to reduce complications.Additionally,it serves as a crucial reference for nurses during the administration of the medication.

Objective To investigate the influencing factors of unplanned rehospitalization within one year after surgery among adult renal transplant recipients. Methods The clinical data of 299 recipients who underwent renal transplant surgery in the Department of Organ Transplantation of the Affiliated Hospital of Guizhou Medical University from January 2020 to December 2022 were retrospectively analyzed. The recipients were divided into unplanned rehospitalization group and non-rehospitalization group based on whether they experienced unplanned rehospitalization within one year after surgery. Univariate analysis and binary Logistic regression analysis were performed to explore the influencing factors of unplanned rehospitalization within one year after renal transplantation. Results Among the 299 recipients, 102 experienced unplanned rehospitalization, with an incidence rate of 34.11%. Univariate analysis revealedstatistically significant differences were noted between the two groups in terms of gender, occupational status, preoperative underlying disease, rejection reactions, nosocomial infections, immunosuppressive medication regimens, serum creatinine, cystatin C, serum phosphorus, serum potassium, and initial hospitalization duration (P < 0.05). Binary Logistic regression analysis showed that having ≥3 kinds of preoperative underlying disease (OR=2.122, 95%CI, 1.198 to 3.759) and experiencing rejection reactions (OR=3.162, 95%CI, 1.217 to 8.218) were independent risk factors for unplanned rehospitalization within one year after renal transplantation (P < 0.05). Conclusion The rate of unplanned rehospitalization within one year after surgery is relatively high among adult renal transplant recipients. Having ≥3 preoperative underlying disease and experiencing rejection reactions are independent risk factors for unplanned rehospitalization within one year after surgery. Transplantation healthcare teams can develop corresponding strategies targeting these risk factors to reduce the rate of unplanned rehospitalization and alleviate the burden of disease.

Objective:To investigate the characteristics of cardiac structure and function changes in maintenance hemodialysis (MHD) patients.Methods:The information of 363 MHD patients with dialysis age ≥3 months who were registered in Huadu District People′s Hospital of Guangzhou City before January 2020 was collected, and the echocardiographic screening was performed to analyze the changes in cardiac structure and function.Results:The most common abnormal changes of heart structure and function in MHD patients were valve regentation (69.7%), left ventricular hypertrophy (LVH) (51.8%), left ventricular diastolic dysfunction (29.8%), valve calcification (11.6%), and left ventricular systolic dysfunction (10.2%). With the increase of age, the left ventricular ejection fraction of MHD patients decreased, and the proportion of left ventricular diastolic dysfunction, left atrial enlargement and pulmonary hypertension increased (all P<0.05). Among 363 MHD patients, 188(51.8%) had LVH. It was found that LVH patients had higher average single ultrafiltration volume, higher brain natriuretic peptide (BNP) level, more type 2 diabetes, lower left ventricular ejection fraction (all P<0.05), and were more prone to chest tightness, chest pain, post activity shortness of breath, heart failure and other symptoms compared with the non-LVH patients (all P<0.05). Conclusions:Most MHD patients have different degrees of cardiac structural changes. Early intervention, reduction of single ultrafiltration volume and control of blood glucose are beneficial to improve the dialysis quality of MHD patients and reduce the occurrence of cardiovascular events.

Objective:To explore the value of cardiac MRI (CMRI) in evaluating left atrial function in patients with postoperative tetralogy of Fallot (rTOF) and postoperative pulmonary stenosis (rPS).Methods:Totally 67 pediatric patients (49 with rTOF, 18 with rPS) with preserved left ventricular ejection fraction (EF) were recruited between January 2019 and October 2021 in Shanghai Children′s Medical Center, School of Medicine, Shanghai Jiao Tong University. Thirty-three healthy volunteers, matched in gender and age, were included as controls from July 2017 to August 2018. Left atrial EF, strain and strain rate of three phases (reservoir, conduit and pump), left atrial volume (maximum volume index, minimum volume index and pre-atrial contraction volume index) were measured with corresponding cardiac function analysis software. Then, the differences in these parameters were analyzed between the three groups by ANOVA or Kruskal-Wallis test with post hoc comparison and Bonferroni correction.Results:Compared with controls, patients with rTOF had lower reservoir function parameters (EF, strain and strain rate), conduit EF, conduit strain, and left atrial maximum volume index ( P<0.05), but higher pump EF ( P<0.05). In patients with rPS, only the reservoir strain rate decreased compared with controls ( P<0.05), and the remaining data showed no significant difference ( P>0.05). The reservoir and conduit EF and strain in patients with rPS were higher than those in patients with rTOF ( P<0.05). Conclusions:In patients with rTOF and rPS, left atrial function has changed despite the preservation of left ventricular EF, which may be an early marker of left ventricular diastolic dysfunction. In children with rTOF, left atrial reservoir and conduit functions decreased while the pump function increased. The reservoir and pump functions in rPS were better than those in rTOF. In addition, CMRI can detect left atrial dysfunction early before it enlarged.

Most literature defines knee extension lag as the failure of the knee joint to achieve the passive extension angle, and some literature defines the inability of the knee to complete the last 15° extension as knee extension lag. Extension lag is one of the common complications after knee surgery, which seriously affects the postoperative knee joint function of patients. Disuse atrophy due to prolonged postoperative bed rest, arthrogenic muscle inhibition due to pain and (or) swelling, direct injury to the knee extensor mechanism during surgery and nerve dysfunction of the quadriceps muscle can all contribute to extension lag. While rehabilitation exercises, physical therapy and medication can significantly improve the patient's symptoms, extension lag due to the knee extension device injury requires surgery if necessary to regain normal knee function. As surgical techniques continue to be refined and rehabilitation interventions advance, the prevention of extension lag and the minimization of angle of knee extension lag need to be given adequate attention by patients, physicians, and researchers. This article reviews clinical studies related to extension lag after knee surgery, focusing on etiology, treatment, prevention and prognosis, with the aim of providing a basis and guidance for the current clinical management.