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 school tuberculosis (TB) outbreaks and preventive treatment in Hefei from 2022 to 2024, so as to provide reference for TB prevention and control in schools. Methods: Data were collected on all school based TB outbreaks occurring during 2022-2024 in Hefei, defined as ≥2 epidemiologically linked TB cases within the same school during a single semester. Statistical analyses were performed using the Chi square test. Results: Close contacts exhibited significantly higher TB incidence (2.88%) and latent mycobacterium tuberculosis infection (LTBI) rates (13.80%) in the school TB outbreaks, compared to non close contacts (0.12% and 2.63%, respectively). Among close contacts, secondary school students showed lower TB incidence (0.48%) and LTBI prevalence (3.42%) than both primary school or younger children (0.68%, 6.95%) and college students ( 0.78% , 6.50%), with statistically significant differences ( χ 2=360.91, 6.37; 791.71, 102.03, all P <0.05). The proportion of LTBI individuals recommended for preventive therapy was higher in primary school or younger groups (98.59%) than in secondary (95.25%) or college students (86.34%) ( χ 2=25.86, P <0.01). However, among those recommended, close contacts had higher uptake (85.82%) and completion rates (87.25%) of preventive therapy than non close contacts (69.63% and 70.57%); similarly, secondary school students demonstrated higher uptake (91.21%) and completion rates (86.45%) compared to primary school or younger (88.57%, 83.87%) and college students (57.28%, 64.08%) ( χ 2=30.52, 26.72; 125.17, 38.84, all P <0.01). Subsequent TB incidence among LTBI close contacts (13.30%) and among those who did not complete preventive therapy (22.73%) were significantly higher than among non close contacts (2.80%, 2.41%), respectively ( χ 2=32.19, 13.87, both P <0.05). Conclusions In school TB outbreaks, close contacts face higher LTBI prevalence and subsequent TB risk than non close contacts. College students show notably low adherence to preventive therapy. It is necessary to take targeted measures to improve the compliance of preventive measures among students.

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.

The development status of the key technologies of first-aid robotics was introduced in autonomous system,embo-died intelligence,digital twins,large artificial intelligence model and autonomous unmanned medical treatment.The present situation in first-aid robotics equipment was reviewed for first-aid diagnosis,treatment,assistance and transportation.The development trends of the key technology and equipment of first-aid robotics were analyzed.It was pointed out the involve-ment of big model-based embodied intelligence technology and digital twins technology in first aid might provide new pers-pectives for the application and advancement of specialized first-aid robotics.[Chinese Medical Equipment Journal,2025,46(3):96-114]

Objective:To compare the efficacy and safety of crisaborole ointment 2% versus pimecrolimus cream 1% in the treatment of mild to moderate atopic dermatitis in children aged 2 years or older.Methods:A multicenter, randomized, open-label, controlled clinical trial was conducted. A total of 120 pediatric patients aged 2 - 17 years with mild to moderate atopic dermatitis were enrolled from departments of dermatology of 8 hospitals in China between March 2022 and February 2023. The participants were randomly assigned in a 1∶1 ratio to the crisaborole group and the pimecrolimus group, and received the treatment with crisaborole ointment 2% and pimecrolimus cream 1% respectively, twice a day for 4 weeks. Visits were scheduled at baseline/on day 1, as well as on days 8, 15, and 29. The primary efficacy outcome was the percentage of patients achieving the Investigator's Static Global Assessment (ISGA) success (defined as clear [0] or almost clear [1] on the ISGA scale, combined with ≥ 2‐grade improvement from baseline) on day 29. The secondary efficacy outcomes included changes in the Eczema Area and Severity Index (EASI) total scores from baseline to day 29, percentages of patients achieving ISGA improvement (defined as clear [0] or almost clear [1] on the ISGA scale), as well as changes in the Peak Pruritus Numerical Rating Scale (NRS) scores, Dermatology Life Quality Index (DLQI) /Infants' Dermatology Life Quality Index (IDLQI) /Children's Dermatology Life Quality Index (CDLQI) scores, and in the Dermatitis Family Impact (DFI) scores. Drug safety was evaluated according to the incidence of adverse events. Categorical data were compared using the chi-square test. Since measurement data did not follow a normal distribution, the rank sum test was used for comparisons of measurement data between groups.Results:A total of 106 children with mild to moderate atopic dermatitis were included in the per-protocol analysis set, with 52 in the crisaborole group (26 males and 26 females) and 54 in the pimecrolimus group (27 males and 27 females). There were no significant differences in age, disease duration, ISGA and EASI scores at baseline between the two groups (all P > 0.05). On day 29, 22 patients (42.31%) in the crisaborole group and 25 (46.30%) in the pimecrolimus group achieved ISGA success, with no significant difference between the two groups ( χ2 = 0.17, P = 0.68) ; 35 patients (67.31%) in the crisaborole group and 45 (83.33%) in the pimecrolimus group achieved ISGA improvement, also with no significant difference between the two groups ( χ2 = 3.68, P = 0.06) ; additionally, there were no significant differences in the EASI, pruritus NRS, DLQI/IDLQI/CDLQI, or DFI scores between the two groups (all P > 0.05). Adverse reactions to the two topical agents were mainly local reactions such as mild to moderate pain, itching, or worsening of itching, and no obvious systemic adverse reactions occurred. The incidence of drug-related adverse reactions was 46.15% (24 cases) in the crisaborole group and 37.04% (20 cases) in the pimecrolimus group, with no significant difference between the two groups ( χ2 = 0.91, P = 0.34) . Conclusion:The efficacy of crisaborole ointment 2% was comparable to that of pimecrolimus cream 1% in the treatment of mild to moderate atopic dermatitis in children aged ≥ 2 years, and it yielded early and rapid improvement in the quality of life of patients and their families, with good safety and tolerability profiles.

BACKGROUND: Based on the China-VHD database, this study sought to develop and validate a Valvular Heart Disease- specific Age-adjusted Comorbidity Index (VHD-ACI) for predicting mortality risk in patients with VHD. METHODS & RESULTS: The China-VHD study was a nationwide, multi-centre multi-centre cohort study enrolling 13,917 patients with moderate or severe VHD across 46 medical centres in China between April-June 2018. After excluding cases with missing key variables, 11,459 patients were retained for final analysis. The primary endpoint was 2-year all-cause mortality, with 941 deaths (10.0%) observed during follow-up. The VHD-ACI was derived after identifying 13 independent mortality predictors: cardiomyopathy, myocardial infarction, chronic obstructive pulmonary disease, pulmonary artery hypertension, low body weight, anaemia, hypoalbuminaemia, renal insufficiency, moderate/severe hepatic dysfunction, heart failure, cancer, NYHA functional class and age. The index exhibited good discrimination (AUC, 0.79) and calibration (Brier score, 0.062) in the total cohort, outperforming both EuroSCORE II and ACCI (P < 0.001 for comparison). Internal validation through 100 bootstrap iterations yielded a C statistic of 0.694 (95% CI: 0.665-0.723) for 2-year mortality prediction. VHD-ACI scores, as a continuous variable (VHD-ACI score: adjusted HR (95% CI): 1.263 (1.245-1.282), P < 0.001) or categorized using thresholds determined by the Yoden index (VHD-ACI ≥ 9 vs. < 9, adjusted HR (95% CI): 6.216 (5.378-7.184), P < 0.001), were independently associated with mortality. The prognostic performance remained consistent across all VHD subtypes (aortic stenosis, aortic regurgitation, mitral stenosis, mitral regurgitation, tricuspid valve disease, mixed aortic/mitral valve disease and multiple VHD), and clinical subgroups stratified by therapeutic strategy, LVEF status (preserved vs. reduced), disease severity and etiology. CONCLUSION The VHD-ACI is a simple 13-comorbidity algorithm for the prediction of mortality in VHD patients and providing a simple and rapid tool for risk stratification.

This consensus will introduce the characteristics of fillers used in the surgical cavities of domestic nasal surgery patients based on relevant literature and expert opinions. It will also provide recommendations for the selection of cavity fillers for different nasal diseases, with chronic sinusitis as a representative example.
Humans ; Nasal Cavity/surgery* ; Nasal Surgical Procedures ; China ; Consensus ; Sinusitis/surgery* ; Dermal Fillers

Otitis media is an infection of the middle ear mainly caused by bacteria, and current treatments rely heavily on antibiotics. However, the emergence of antibiotic-resistant bacterial strains seriously affects their efficacy. In our study, we found that extracellular vesicles (EVs) derived from human natural killer cells (NKs) inhibit the proliferation of both standard and levofloxacin (LVX)-resistant strains of Staphylococcus aureus in a dose-dependent manner. Moreover, compared to LVX, EVs were more effective at reducing effusion and rescuing hearing thresholds in animal models. For LVX-sensitive strains, EVs were significantly more effective in terms of curative time but not curative rate. For LVX-resistant strains, EVs were significantly more effective in terms of both curative rate and curative time when applied alone or applied jointly with LVX. In summary, we found that NK EVs are highly effective in treating otitis media, providing an alternative approach for treating this common disease.
Killer Cells, Natural/metabolism* ; Exosomes/metabolism* ; Animals ; Humans ; Otitis Media/therapy* ; Staphylococcus aureus/drug effects* ; Disease Models, Animal ; Anti-Bacterial Agents/pharmacology* ; Levofloxacin/pharmacology*

OBJECTIVE: Human brucellosis is a serious public health concern in the Xilingol League, Inner Mongolia; however, the epidemic trends are unclear. METHOD: In this study, Joinpoint regression analysis and spatiotemporal analysis were applied to investigate the epidemic evolution of human brucellosis. RESULT: From 2004 to 2023, a total of 35,747 cases were reported, with an annual average of 1787.35 cases and an annual average incidence rate of 176.04/100,000. The incidence increased from 173.96/100,000 in 2004 to 500.71/100,000 in 2009 and fluctuated to 61.43/100,000 in 2023. Three epidemic join points were observed in which the disease experienced an alternative rise and fall, peaking in 2009 (APC = 21.73, P > 0.001) and 2020 (APC = 21.51, P > 0.001). The disease showed a persistent decline trend in lentitude (AAPC = -5.30, P > 0.001), suggesting challenges in disease control and a higher risk of rebound. The most cases were reported in Xilinhot City ( n = 4,777), followed by 4,391 in Sonid Left Banner, and 4,324 in Abaga Banner. Spatiotemporal analysis revealed two high clusters (CI and CII) from 2005 to 2012, the high cluster encompassing eight counties and shifting from north to south. CONCLUSION The present analysis highlights that human brucellosis has decreased significantly in the Xilingol League, but the epidemic is still severe; further implementation of a strict control program is necessary.
China/epidemiology* ; Humans ; Brucellosis/epidemiology* ; Epidemics ; Spatio-Temporal Analysis ; Incidence ; Cluster Analysis

OBJECTIVE: Recombination events are common and serve as the primary driving force of diverse human adenovirus (HAdV), particularly in children with acute respiratory tract infections (ARIs). Therefore, continual monitoring of these events is essential for effective viral surveillance and control. METHODS: Respiratory specimens were collected from children with ARIs between January 2022 and December 2023. The penton base, hexon, and fiber genes were amplified from HAdV-positive specimens and sequenced to determine the virus type. In cases with inconsistent typing results, genes were cloned into the pGEM-T vector to detect recombination events. Metagenomic next-generation sequencing (mNGS) was performed to characterize the recombinant HAdV genomes. RESULTS: Among 6,771 specimens, 277 (4.09%, 277/6,771) were positvie for HAdV, of which 157 (56.68%, 157/277) were successfully typed, with HAdV-B3 being the dominant type (91.08%, 143/157), and 14 (5.05%, 14/277) exhibited inconsistent typing results, six of which belonged to species B. The penton base genes of these six specimens were classified as HAdV-B7, whereas their hexon and fiber genes were classified as HAdV-B3, resulting in a recombinant genotype designated P7H3F3, which closely resembled HAdV-B114. Additionally, a partial gene encoding L1 52/55 kD was identified, which originated from HAdV-B16. CONCLUSION A novel recombinant, P7H3F3, was identified, containing sequences derived from HAdV-B3 and HAdV-B7, which is similar to HAdV-B114, along with additional sequences from HAdV-B16.
Humans ; Adenoviruses, Human/isolation & purification* ; Respiratory Tract Infections/epidemiology* ; Child, Preschool ; Child ; Recombination, Genetic ; Male ; Beijing/epidemiology* ; Infant ; Female ; Phylogeny ; Adenovirus Infections, Human/epidemiology* ; Acute Disease ; Genome, Viral