Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Background Asthma poses a serious threat to children's growth, development, and mental health, thus there has been an increasing focus on the control of asthma morbidity in children and the assessment of its risk factors. A growing body of research has found that exposure to ambient air pollutants an significatly increase the risk of childhood asthma. Objective To understand the changes of ambient air pollutant concentrations in Nanjing and asthma outpatient visits to Nanjing Children's Hospital, and to quantitatively analyze the effects of exposure to different ambient air pollutants on children's asthma outpatient visits. Methods Daily data of ambient air pollutants fine particulate matter (PM_2.5), inhalable particle (PM₁₀), sulfur dioxide (SO₂), nitrogen dioxide (NO₂), carbon monoxide (CO), ozone (O₃), meteorological factors (air temperature & relative humidity), and outpatient visits due to asthma in the hospital from January 1, 2019 to December 31, 2023 were collected, and a generalized additive model based on quasi poisson distributions was used to quantitatively analyze the short-term effects of ambient air pollutant exposure on outpatient visits due to asthma in the hospital. Results The annual average concentrations of PM_2.5, PM₁₀, SO₂, and NO₂ in Nanjing from 2019 to 2023 did not exceed the national limits. For single-day lagged effects, the single-pollutant model showed that the effects of PM_2.5, PM₁₀, NO₂, and CO on children's asthma outpatient visits were greatest for every 10 units increase at lag0, with excess risk (ER) of 1.39% (95%CI: 0.65%, 2.14%), 1.46% (95%CI: 0.97%, 1.95%), 5.46% (95%CI: 4.36%, 6.57%), and 0.18% (95%CI: 0.11%, 0.26%), respectively, and SO₂ reached the maximum effect at lag1, with an ER of 23.15% (95%CI: 13.57%, 33.53%) for each 10 units increase in concentration. Different pollutants reached their maximum cumulative lag effects at different time. The PM₁₀, PM_2.5, SO₂, NO₂, and CO showed the largest cumulative lag effects at lag01, lag01, lag02, lag02, and lag03, respectively, with ERs of 1.35% (95%CI: 0.77%, 1.92%), 0.96% (95%CI: 0.10%, 1.83%), 28.50% (95%CI: 15.49%, 42.98%), 6.92% (95%CI: 5.53%, 8.33%), and 0.31% (95%CI: 0.20%, 0.42%), respectively. The influences of PM_2.5 and PM₁₀ on outpatient visits due to asthma in the hospital became more pronounced with advancing age, while the associations with NO₂, SO₂, and CO were weakened as children grew older. Conclusion Ambient air pollutants (PM_2.5, PM₁₀, SO₂, NO₂, CO) can increase childhood asthma visits, and different pollutants have varied effects on the number of asthmatic children's visits at different ages.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

OBJECTIVE: To observe the efficacy and safety of 21-day venetoclax (VEN) plus azacitidine (AZA) (21-day VA) in newly diagnosed unfit acute myeloid leukemia (AML) patients in the real-world. METHODS: The clinical data of patients with unfit-AML who received 21-day VA regimen from December 2020 to July 2024 in our center and completed at least 1 cycle of therapeutic effect assessment was retrospectively collected to analyze the safety, efficacy and its influencing factors. RESULTS: A total of 59 patients were enrolled in our study, with a median age of 67(48-87) years old. After 1 cycle of therapy, the composite complete remission (cCR) rate was 74.5%, 54.2% of cases were negative for minimal residual disease (MRD). Among them, the MRD negative rate of patients with NPM1 mutation was significantly higher than that of patients without NPM1 mutation ( P =0.032). The median follow-up of patients was 19(2-38) months, the best cCR and MRD negative rates were 78% and 64.4%, respectively, the median overall survival (OS) time was 12 months, and the median progression free survival (PFS) time was 5 months. Multivariate Cox regression analysis showed less than 4 cycles of VA chemotherapy were independent risk factor for PFS and OS ( P < 0.05). After achieving remission, anemia and thrombocytopenia improved with the increase of the number of chemotherapy cycle. CONCLUSION In real-world, 21-day VA regimen still shows significant efficacy in the treatment of newly diagnosed unfit-AML, without adversely affecting remission rate and MRD negative rate of the first cycle.
Humans ; Leukemia, Myeloid, Acute/drug therapy* ; Aged ; Middle Aged ; Bridged Bicyclo Compounds, Heterocyclic/therapeutic use* ; Sulfonamides/therapeutic use* ; Azacitidine/therapeutic use* ; Aged, 80 and over ; Male ; Female ; Retrospective Studies ; Nucleophosmin ; Antineoplastic Combined Chemotherapy Protocols/therapeutic use* ; Remission Induction ; Mutation ; Treatment Outcome

Sustained inflammatory responses are closely related to various severe diseases, and inhibiting the excessive activation of inflammasomes and pyroptosis has significant implications for clinical treatment. Natural products have garnered considerable concern for the treatment of inflammation. Huanglian-Wumei decoction (HLWMD) is a classic prescription used for treating inflammatory diseases, but the necessity of their combination and the exact underlying anti-inflammatory mechanism have not yet been elucidated. Inspired by the supramolecular self-assembly strategy and natural drug compatibility theory, we successfully obtained berberine (BBR)-chlorogenic acid (CGA) supramolecular (BCS), which is an herbal pair from HLWMD. Using a series of characterization methods, we confirmed the self-assembly mechanism of BCS. BBR and CGA were self-assembled and stacked into amphiphilic spherical supramolecules in a 2:1 molar ratio, driven by electrostatic interactions, hydrophobic interactions, and π-π stacking; the hydrophilic fragments of CGA were outside, and the hydrophobic fragments of BBR were inside. This stacking pattern significantly improved the anti-inflammatory performance of BCS compared with that of single free molecules. Compared with free molecules, BCS significantly attenuated the release of multiple inflammatory mediators and lipopolysaccharide (LPS)-induced pyroptosis. Its anti-inflammatory mechanism is closely related to the inhibition of intracellular nuclear factor-kappaB (NF-κB) p65 phosphorylation and the noncanonical pyroptosis signalling pathway mediated by caspase-11.

Nutritional support therapy is one of the extremely important treatment methods for patients in the intensive care unit. Timely and effective nutritional support regimens can improve patients' immune function, reduce complications, and optimize clinical outcomes. Energy expenditure is influenced by multiple factors, including patients' baseline characteristics (such as physical condition, gender, age) and dynamic changes in indicators (such as body temperature, nutritional support regimens, and therapeutic interventions). The currently recognized "gold standard" for accurately assessing energy metabolism in clinical practice is the indirect calorimetry system, also known as the metabolic cart. This device monitors carbon dioxide production and oxygen consumption in real time and uses specific algorithms to estimate the metabolic proportions of the three major nutrients (carbohydrates, fats, and proteins) in energy expenditure. An appropriate nutrient ratio helps maintain the balance between supply and demand in the body's nutritional metabolism. In the management of critically ill patients, the application of the metabolic cart enables personalized nutritional therapy, avoiding over- or under-supply of energy and optimizing the use of medical resources. Furthermore, with real-time, quantitative data support from the energy metabolism monitoring system, clinicians can develop more precise nutritional intervention strategies, thereby improving patient prognosis. This article provides a systematic review of the technical features of the metabolic cart and its application value in various critical care scenarios, aiming to offer a reference for indirect calorimetry in clinical practice.
Humans ; Critical Illness/therapy* ; Nutritional Support ; Energy Metabolism ; Calorimetry, Indirect

OBJECTIVE: This study reports the first imported case of Lassa fever (LF) in China. Laboratory detection and molecular epidemiological analysis of the Lassa virus (LASV) from this case offer valuable insights for the prevention and control of LF. METHODS: Samples of cerebrospinal fluid (CSF), blood, urine, saliva, and environmental materials were collected from the patient and their close contacts for LASV nucleotide detection. Whole-genome sequencing was performed on positive samples to analyze the genetic characteristics of the virus. RESULTS: LASV was detected in the patient's CSF, blood, and urine, while all samples from close contacts and the environment tested negative. The virus belongs to the lineage IV strain and shares the highest homology with strains from Sierra Leone. The variability in the glycoprotein complex (GPC) among different strains ranged from 3.9% to 15.1%, higher than previously reported for the seven known lineages. Amino acid mutation analysis revealed multiple mutations within the GPC immunogenic epitopes, increasing strain diversity and potentially impacting immune response. CONCLUSION The case was confirmed through nucleotide detection, with no evidence of secondary transmission or viral spread. The LASV strain identified belongs to lineage IV, with broader GPC variability than previously reported. Mutations in the immune-related sites of GPC may affect immune responses, necessitating heightened vigilance regarding the virus.
Humans ; China/epidemiology* ; Genome, Viral ; Lassa Fever/virology* ; Lassa virus/classification* ; Molecular Epidemiology ; Phylogeny