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: 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.

The study was conducted to investigate the mechanism of Xinyang Tablets( XYP) in modulating the fat mass and obesity-associated protein(FTO)/N6-methyladenosine(m6A) signaling pathway to ameliorate ventricular remodeling in heart failure(HF). A mouse model of HF was established by transverse aortic constriction(TAC). Mice were randomized into sham, model, XYP(low, medium, and high doses), and positive control( perindopril) groups(n= 10). From day 3 post-surgery, mice were administrated with corresponding drugs by gavage for 6 consecutive weeks. Following the treatment, echocardiography was employed to evaluate the cardiac function, and RT-qPCR was employed to determine the relative m RNA levels of key markers, including atrial natriuretic peptide( ANP), B-type natriuretic peptide( BNP), β-myosin heavy chain(β-MHC), collagen type I alpha chain(Col1α), collagen type Ⅲ alpha chain(Col3α), alpha smooth muscle actin(α-SMA), and FTO. The cardiac tissue was stained with Masson's trichrome and wheat germ agglutinin(WGA) to reveal the pathological changes. Immunohistochemistry was employed to detect the expression levels of Col1α, Col3α, α-SMA, and FTO in the myocardial tissue. The m6A modification level in the myocardial tissue was measured by the m6A assay kit. An H9c2 cell model of cardiomyocyte injury was induced by angiotensin Ⅱ(AngⅡ), and small interfering RNA(siRNA) was employed to knock down FTO expression. RT-qPCR was conducted to assess the relative m RNA levels of FTO and other genes associated with cardiac remodeling. The m6A modification level was measured by the m6A assay kit, and Western blot was employed to determine the phosphorylated phosphatidylinositol 3-kinase(p-PI3K)/phosphatidylinositol 3-kinase(PI3K) and phosphorylated serine/threonine kinase(p-Akt)/serine/threonine kinase(Akt) ratios in cardiomyocytes. The results of animal experiments showed that the XYP treatment significantly improved the cardiac function, reduced fibrosis, up-regulated the m RNA and protein levels of FTO, and lowered the m6A modification level compared with the model group. The results of cell experiments showed that the XYP-containing serum markedly up-regulated the m RNA level of FTO while decreasing the m6A modification level and the p-PI3K/PI3K and p-Akt/Akt ratios in cardiomyocytes. Furthermore, FTO knockdown reversed the protective effects of XYP-containing serum on Ang Ⅱ-induced cardiomyocyte hypertrophy. In conclusion, XYP may ameliorate ventricular remodeling by regulating the FTO/m6A axis, thereby inhibiting the activation of the PI3K/Akt signaling pathway.
Animals ; Ventricular Remodeling/drug effects* ; Heart Failure/physiopathology* ; Signal Transduction/drug effects* ; Mice ; Male ; Alpha-Ketoglutarate-Dependent Dioxygenase FTO/genetics* ; Drugs, Chinese Herbal/administration & dosage* ; Mice, Inbred C57BL ; Humans ; Adenosine/analogs & derivatives* ; Myocytes, Cardiac/metabolism* ; Disease Models, Animal

This study investigates the effect and underlying mechanism of Guizhi Tongluo Tablets(GZTL) in treating atherosclerosis(AS) in a mouse model. Apolipoprotein E-knockout(ApoE~(-/-)) mice were randomly assigned to the following groups: model, high-, medium-, and low-dose GZTL, and atorvastatin(ATV), and age-matched C57BL/6J mice were selected as the control group. ApoE~(-/-) mice in other groups except the control group were fed with a high-fat diet for the modeling of AS and administrated with corresponding drugs via gavage for 8 weeks. General conditions, signs of blood stasis, and body mass of mice were monitored. Aortic plaques and their stability were assessed by hematoxylin-eosin, Masson, and oil red O staining. Serum levels of total cholesterol(TC), triglycerides(TG), and low-density lipoprotein cholesterol(LDL-C) were measured by biochemical assays, and those of interleukin-1β(IL-1β), tumor necrosis factor-α(TNF-α), and interleukin-6(IL-6) were determined via enzyme-linked immunosorbent assay. Apoptosis was assessed by terminal deoxynucleotidyl transferase dUTP nick end labeling(TUNEL). Single-cell RNA sequencing(scRNA-seq) was employed to analyze the differential expression of CD72hi macrophages(CD72hi-Mφ) in the aortas of AS patients and mice. The immunofluorescence assay was employed to visualize CD72hi-Mφ expression in mouse aortic plaques, and real-time fluorescence quantitative PCR was utilized to determine the mRNA levels of IL-1β, TNF-α, and IL-6 in the aorta. The results demonstrated that compared with the control group, the model group exhibited significant increases in body mass, aortic plaque area proportion, necrotic core area proportion, and lipid deposition, a notable decrease in collagen fiber content, and an increase in apoptosis. Additionally, the model group showcased elevated serum levels of TC, TG, LDL-C, IL-1β, TNF-α, and IL-6, alongside marked upregulations in the mRNA levels of IL-1β, TNF-α, and IL-6 in the aorta. In comparison with the model group, the GZTL groups and the ATV group showed a reduction in body mass, and the medium-and high-dose GZTL groups and the ATV group demonstrated reductions in aortic plaque area proportion, necrotic core area proportion, and lipid deposition, an increase in collagen fiber content, and a decrease in apoptosis. Furthermore, the treatment goups showcased lowered serum levels of TC, TG, LDL-C, IL-1β, TNF-α, and IL-6. The data of scRNA-seq revealed significantly elevated CD72hi-Mφ signaling in carotid plaques of AS patients compared with that in the normal arterial tissue. Animal experiments confirmed that CD72hi-Mφ expression, along with several pro-inflammatory cytokines, was significantly upregulated in the aortas of AS mice, which were downregulated by GZTL treatment. In conclusion, GZTL may alleviate AS by inhibiting CD72hi-Mφ activity.
Animals ; Drugs, Chinese Herbal/administration & dosage* ; Atherosclerosis/immunology* ; Mice ; Mice, Inbred C57BL ; Macrophages/immunology* ; Male ; Humans ; Apolipoproteins E/genetics* ; Tablets ; Tumor Necrosis Factor-alpha/genetics* ; Apoptosis/drug effects* ; Interleukin-1beta/genetics* ; Interleukin-6/genetics* ; Disease Models, Animal ; Mice, Knockout

Syringa pinnatifolia, belonging to the family Oleaceae, is a species endemic to China. It is predominantly distributed in the Helan Mountains region of Inner Mongolia and Ningxia of China. The peeled roots, stems, and thick branches have been used as a distinctive Mongolian medicinal material known as "Shan-chen-xiang", which has effects such as suppressing "khii", clearing heat, and relieving pain and is employed for the treatment of cardiovascular and pulmonary diseases and joint pain. Over the past five years, significant increase was achieved in research on chemical constituents and pharmacological effects. There were a total of 130 new constituents reported, covering sesquiterpenoids, lignans, and alkaloids. Its effects of anti-myocardial ischemia, anti-cerebral ischemia/reperfusion, sedation, and analgesia were revealed, and the mechanisms of agarwood formation were also investigated. To better understand its medical value and potential of clinical application, this review updates the research progress in recent five years focusing on the chemical constituents and pharmacological effects of S. pinnatifolia, providing reference for subsequent research on active ingredient and support for its innovative application in modern medicine system.
Medicine, Mongolian Traditional ; Humans ; Drugs, Chinese Herbal/pharmacology* ; Animals ; Syringa/chemistry*

Incomplete debridement of chronic osteomyelitis is the main factor leading to recurrence. For the treatment of chronic osteomyelitis, the complete elimination of the source of infection is the key to preventing recurrence. This process includes not only the complete removal of infected lesions, dead bone, accreted scar tissue and granulation tissue, but also the elimination of dead space and improved local blood circulation. In these steps, debridement is a core procedure, and judging the scope of debridement is the premise of whether it could be completely debridement. This article systematically reviewed the application of different imaging techniques in evaluating the scope of chronic osteomyelitis infection, and discusses its future development trend. Although traditional plain X-ray film could preliminarily indicate osteomyelitis, it is difficult to determine the infection scope. CT scan has the function of accurate anatomic localization, which is important for preoperative assessment of the scope of bone infection, but the recognition of soft tissue information is limited. MRI, with its high sensitivity, clearly distinguishes between infected bone and soft tissue, which plays an important role in the evaluation of soft tissue infection, but may overestimate the extent of bone infection. Nuclide techniques such as 18F-FDG PET/CT and SPECT/CT show great potential for accurately assessing the extent of infection before surgery. In the future, by optimizing the combination of different imaging technologies, combining clinical symptoms, intraoperative conditions and pathological results, and developing an image analysis platform based on artificial intelligence, it will be able to more accurately assess the scope of infection, provide more effective and personalized treatment plans for patients with chronic osteomyelitis, enhance treatment effects, and significantly improve quality of life of patients.
Humans ; Osteomyelitis/diagnosis* ; Chronic Disease ; Magnetic Resonance Imaging ; Tomography, X-Ray Computed