OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

OBJECTIVE:Based on different algorithms of machine learning,the prediction model of lumbar disc herniation has become a trend and hot spot in the development of precision medicine.However,there is limited evidence on the reporting quality and methodological quality of prediction models of lumbar disc herniation outcomes using machine learning.This article is aimed to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation by comprehensively analyzing the report quality and risk of bias of previous studies that developed and validated prognosis prediction models based on machine learning through a comprehensive literature search,in order to explore the performance of machine learning algorithms in predicting the prognosis of lumbar disc herniation.METHODS:The databases of CNKI,WanFang,VIP,SinOMED,PubMed,Web of Science,Embase,and The Cochrane Library were searched by computer.Studies on the use of machine learning to develop(and/or validate)prognostic prediction models for lumbar disc herniation were collected from the inception of the database to December 31,2023.Two researchers independently screened the literature,extracted data,and assessed the risk of bias of the included studies.The reporting quality and risk of bias of the included studies were assessed by the Multivariable Transparent Reporting of Predictive Models(TRIPOD)statement and the Predictive Model Risk of Bias Assessment Tool(PROBAST).The results of the evaluation were analyzed using descriptive statistics and visual charts.RESULTS:(1)A total of 23 articles were included,and the TRIPOD compliance of each study ranged from 11％to 87％,with a median compliance of 54％.The quality of reporting of titles,detailed descriptions of treatment measures,blinding of predictors,handling of missing data,details of risk stratification,specific procedures for enrollment,model interpretation,and model performance was mostly poor,with TRIPOD adherence rates ranging from 4％to 35％.(2)Of all included studies,61％had a high risk of bias and 39％had an unclear overall risk of bias.The area under the curve,accuracy,sensitivity and specificity were used to evaluate the performance of the model.The areas under the curve of 20 models were reported,ranging from 0.561 to 0.999.Three models reported the accuracy of the model,ranging from 82.07％to 89.65％.(3)Among all included studies,the statistical analysis domain was most often assessed as having a high risk of bias,mainly due to the small number of valid samples,the selection of predictors based on univariate analysis and the lack of calibration and discrimination assessment of the model in the study.CONCLUSION:These results indicate that machine learning can achieve good predictive ability in the development and validation of prognostic models for lumbar disc herniation.The commonly used algorithms include regression algorithm,support vector machine,decision tree,random forest,artificial neural network,naive Bayes and other algorithms.Reasonable algorithms combined with clinical practice can improve the accuracy of prognosis prediction of lumbar disc herniation.However,the reporting and methodological quality of prognosis prediction models based on machine learning are poor,the prediction performance of different models varies greatly,and the generalization and extrapolation of research models are unclear.There is an urgent need to improve the design,implementation and reporting of such studies.To promote the application of machine learning in the clinical practice of lumbar disc herniation prediction models,it is necessary to comprehensively consider various predictors related to the prognosis of the disease before modeling,and strictly follow the relevant standards of PROBAST tool during modeling.

Objective To analyze the clinical and therapeutic characteristics of Epstein-Barr virus (EBV)-negative posttransplant lymphoproliferative disease (PTLD) with diffuse large B-cell lymphoma (DLBCL) in the context of specific cases and literature. Methods A case of EBV-negative DLBCL (GCB type) after kidney transplantation is reported. The patient was a 45-year-old male who underwent living-related kidney transplantation in 2016 and has been receiving triple immunosuppressive therapy with tacrolimus, mycophenolate mofetil and methylprednisolone since then. In 2024, the patient presented with intermittent fever, night sweats and gastrointestinal symptoms. The diagnosis was confirmed by endoscopic pathology, immunohistochemical staining and positron emission tomography/computed tomography. The R-CDOP regimen (rituximab + cyclophosphamide + liposomal doxorubicin + vincristine + dexamethasone) was used for treatment. Results The patient was diagnosed with EBV-negative DLBCL (GCB type, Ann Arbor stage Ⅳ B). After 4 cycles of R-CDOP chemotherapy, the efficacy assessment was partial remission, and the transplant kidney function remained stable. Conclusions For EBV-negative PTLD after kidney transplantation, it is necessary to break through the "virus-dependent" diagnostic thinking. In clinical practice, the focus should be on protecting the transplant kidney, and individualized treatment plans should be developed for patients.

Background: This study was performed to characterize the incidence, costs, and risk factors associated with renal and cardiovascular adverse outcomes following primary percutaneous coronary intervention (pPCI) in patients with ST-elevation myocardial infarction (STEMI). Methods: Patients with STEMI who underwent pPCI using iso-osmolar contrast were enrolled at 39 centers. The incidence of acute kidney injury (AKI) and major adverse renal and cardiovascular events (MARCE) was analyzed, as well as inpatient costs. Logistic regression analysis was performed to identify risk factors. Results: Among 2,293 patients, the incidence of AKI and MARCE within 72 hours post-pPCI was 4.14% (n = 95) and 4.40% (n = 101), respectively. AKI and/or MARCE were associated with systolic blood pressure (AKI: odds ratio [OR], 1.009; 95% confidence interval [CI], 1.000–1.018), hypertension (AKI: OR, 1.815; 95% CI, 1.133–2.906; MARCE: OR, 1.760;95% CI, 1.118–2.769), anterior wall infarction (AKI: OR, 1.895; 95% CI, 1.196–3.004; MARCE:OR, 1.939; 95% CI, 1.240–3.032), Killip class (AKI: OR, 1.465; 95% CI, 1.117–1.922; MARCE:OR, 1.467; 95% CI, 1.131–1.903), and serum creatinine (SCr; MARCE: OR, 1.006; 95% CI, 1.000–1.012). Hospitalization costs for patients with STEMI who developed AKI or MARCE were significantly higher than for those without AKI (9,595 ± 5,795 vs. 8,279 ± 3,872 USD, P = 0.003) or without MARCE (9,890 ± 5,616 vs. 8,255 ± 3,859 USD, P < 0.001). Conclusions In patients with STEMI undergoing pPCI with iso-osmolar contrast, the incidence of AKI and MARCE was associated with higher hospitalization costs. Systolic blood pressure, hypertension, anterior wall infarction, Killip class, and SCr were identified as risk factors for these outcomes.

Effective symptom management can alleviate the physical and psychological distress experienced by patients with cancer, improve quality of life, and contribute to treatment adherence and improve clinical outcomes. However, most existing guidelines are developed for healthcare professionals, and patients and the public have limited access to standardized and comprehensible guidance on symptom management. To address this gap and to facilitate effective communication and shared decision-making, this study proposes the development of a patient version of the cancer symptom management guideline. The development process will adhere to the methodological framework recommended by the Guidelines International Network and the World Health Organization. The GRADE approach will be employed to assess the certainty of evidence and to formulate recommendations. In addition, the process will be informed by the Appraisal of Guidelines for Research and Evaluation Ⅱ (AGREE Ⅱ) instrument and the Reporting Items for Practice Guidelines in Healthcare-Public or Patient Versions of Guidelines (RIGHT-PVG). This protocol outlines the establishment of the guideline working group, the identification and prioritization of key questions, evidence retrieval and appraisal, and the formulation of recommendations, with the aim of ensuring methodological rigor and transparency in the development of the patient guideline and providing methodological reference for similar guideline initiatives.

Objective To evaluate the performance of lightweight Chinese large language models (LLMs) in answering specialized lung cancer questions, and to explore the impact of retrieval-augmented generation (RAG) on model performance. Methods Eleven lightweight Chinese LLMs with parameter sizes ranging from 7B to 32B were included. A lung cancer-specific evaluation dataset consisting of 200 questions [100 A1-type (basic knowledge) and 100 A2-type (clinical case) questions], constructed based on clinical guidelines and thoracic surgery textbooks, was used for assessment. Model performance was evaluated under two conditions (with and without RAG). Accuracy was used to assess model performance, and response latency was recorded to reflect inference efficiency. An accuracy–latency scatter plot was constructed for descriptive analysis of overall model performance. Results All models successfully completed the evaluation. With the introduction of RAG, the overall average accuracy improved from 61.68% to 76.36%. Smaller models demonstrated the most significant improvement (e.g., the accuracy of DeepSeek-7B increased from 32.50% to 60.00%, P<0.001). The average response latency increased from 12.58 s to 13.80 s. The Qwen3 series showed the best overall performance, and Qwen3-32B achieved the highest accuracy under both conditions (76.50% and 84.00%, respectively). After RAG integration, performance differences among model families were markedly reduced. Based on the accuracy-latency trade-off, Qwen3-32B achieved the best balance between accuracy and response latency under the baseline condition, whereas Qwen3-14B demonstrated superior overall performance in terms of accuracy, latency, and computational cost after RAG integration. Conclusion The integration of RAG technology improves the ability of lightweight Chinese LLMs to answer specialized lung cancer questions. Under the dual practical constraints of limited computational resources and medical data security requirements, the "lightweight model+RAG" technical framework may represent a promising deployment solution.

Objective To investigate the association between dietary patterns and adolescent development in girls in Pudong New Area，Shanghai. Methods Using a stratified cluster sampling method, a total of 998 girls were selected from the cohort of growth, development, and health of children and adolescents in Pudong New Area of Shanghai. Physical development examinations and dietary surveys were conducted on these girls. Principal component analysis was used to determine girls' dietary patterns, and their relationship with adolescent development was analyzed using logistic regression. Results Among the 998 participants, 770 (77.15%) were pubertal girls. The factor analysis identified three dietary patterns, fungi and algae~nuts~legumes pattern, meat-vegetables and fruits-cereals pattern, and milk-eggs pattern. Logistic regression analysis revealed that after adjusting for age, waist circumference, residential district, mother's education level, and father's education level, the Q2 level (OR=0.54, 95% CI: 0.30-0.99), Q3 level (OR=0.50, 95% CI: 0.27-0.91), and Q4 level (OR=0.49, 95% CI: 0.26-0.90) of the fungi and algae-nuts-legumes pattern were negatively correlated with adolescent development. The Q2 level (OR=0.48, 95% CI: 0.27-0.87) and Q3 level (OR=0.49, 95% CI: 0.27-0.88) of the milk-eggs pattern were also negatively correlated with adolescent development. Conclusion The fungi and algae~nuts~legumes pattern and milk-eggs patterns are inversely related to pubertal development.

Nitrogen oxides(NOx=NO+NO2)are important precursors of ozone(O3),and NOx and its oxides together constitute reactive nitrogen oxides(NOy)in the atmosphere.A comprehensive understanding of the total NOy level in the atmosphere is of great significance for a deeper understanding of the atmospheric nitrogen cycle and oxidation,as well as for formulating strategies for air pollution prevention and control.In this work,a thermal decomposition-broadband cavity enhanced absorption spectroscopy(TD-BBCEAS)technique for online measurement of total NOy in the atmosphere was developed.With this method,the NOy was efficiently converted into NO2,and the total NOy concentration in the atmosphere was indirectly obtained by measuring NO2.Focusing on the key factors affecting the measurement of total NOy,the influence of NO titration efficiency and other NOy component TD efficiency on measurement accuracy was emphasized.By changing the oxygen(O2)flow rate through the mercury lamp to alter the O3 concentration for titrating NO,the conversion efficiency of NO was evaluated.At O2 flow rate of 6 mL/min,the conversion efficiency of NO was greater than 99%.TD efficiency testing and analysis on NO2,peroxyacetyl nitrate(PAN),nitric acid(HNO3),and nitrous acid(HONO),which account for a large proportion of atmospheric NOy components,was carried out using 680℃as the optimal TD temperature for efficient conversion of NOy.With NO and HONO sample gases as typical verification gases,the conversion efficiency of NOy and the accuracy of NOy measurement by TD-BBCEAS system were verified by switching the on and off modes of mercury lamp and TD device.At integration time of 60 s,the detection limit of the system for NOy was 2.83×1010 molecules/cm3(60 s,2σ).A comparative measurement of actual atmospheric NOy was conducted between the TD-BBCEAS system and the NOy analyzer.The observation results showed a correlation coefficient(R2)of 0.98 and a slope of 0.93,further verifying the feasibility and accuracy of applying the TD-BBCEAS system to measurement of total NOy.

Metal-organic frameworks(MOFs)are a class of organic-inorganic hybrid materials formed by the self-assembly of metal ions or metal clusters with organic ligands through coordination,and possess high specific surface area,tunable pore size and diverse structures.In recent years,MOFs and their composites have shown great application potential in the field of drug detection,especially in selective recognition,enhancing detection sensitivity and on-site rapid detection.This paper summarized the structural characteristics,synthesis methods and detection principles of MOFs and their composites,and reviewed the latest research progresses in detection of various drugs such as opioids,amphetamines,cannabinoids,cathinones,cocaine,ketamine,fentanyls and psychotropic drugs.The advantages and challenges of MOFs materials in the pretreatment of complex biological samples,sensor construction and on-site rapid detection were discussed,and the prospects for future development were analyzed,with the aim of providing theoretical support and technical references for promoting the applications of MOFs in anti-drug practice.

Objective To analyze the risk factors of critically ill patients infected with Acinetobacter bau-mannii(AB)and carbapenem resistant Acinetobacter baumannii(CRAB).Methods From January 2022 to June 2023,the data of Intensive Care Unit(ICU)patients admitted to Second Hospital of Shanxi Medical Uni-versity in Shanxi Province were collected.According to whether they were infected with AB,the patients were divided into an observation group and a control group(98 cases each).The observation group was further di-vided into a carbapenem sensitive Acinetobacter baumannii(CSAB)group(72 cases)and a CRAB group(26 cases).Mann-Whitney U test,chi-square test and other univariate and multivariate binary Logistic regression were used to analyze the risk factors of AB and CRAB infection for critically ill patients.The prognosis was analyzed by Kaplan Meier survival analysis.Results Long stay in ICU,previous use of carbapenem drugs and high Acute Physiology and Health Evaluation(APACHE Ⅱ)score were independent risk factors for AB sus-ceptibility(P＜0.05),while the independent risk factors for CRAB susceptibility were invasive ventilation and delayed surgery(P＜0.01).In addition,CRAB infection,COVID-19 and shock was risk factors for death in critically ill patients,and invasive ventilation,indwelling drainage tube and operation could reduce the risk of death in critically ill patients(P＜0.05).Conclusion ICU stay time,APACHE Ⅱ score,previous use of car-bapenem drugs and invasive ventilation increase the risk of AB and CRAB infection in critically ill patients.In-vasive ventilation,indwelling drainage and early surgery could reduce the risk of death from AB and CRAB in-fection in critically ill patients.