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.

Neuroscience is a significant frontier discipline within the natural sciences and has become an important interdisciplinary frontier scientific field. Brain is one of the most complex organs in the human body, and its structural and functional analysis is considered the “ultimate frontier” of human self-awareness and exploration of nature. Driven by the strategic layout of “China Brain Project”, Chinese scientists have conducted systematic research focusing on “understanding the brain, simulating the brain, and protecting the brain”. They have made breakthrough progress in areas such as the principles of brain cognition, mechanisms and interventions for brain diseases, brain-like computation, and applications of brain-machine intelligence technology, aiming to enhance brain health through biomedical technology and improve the quality of human life. Due to limited understanding and comprehension of neuroscience, there are still many important unresolved issues in the field of neuroscience, resulting in a lack of effective measures to prevent and protect brain health. Therefore, in addition to actively developing new generation drugs, exploring non pharmacological treatment strategies with better health benefits and higher safety is particularly important. Epidemiological data shows that, exercise is not only an indispensable part of daily life but also an important non-pharmacological approach for protecting brain health and preventing neurodegenerative diseases, forming an emerging research field known as motor neuroscience. Basic research in motor neuroscience primarily focuses on analyzing the dynamic coding mechanisms of neural circuits involved in motor control, breakthroughs in motor neuroscience research depend on the construction of dynamic monitoring systems across temporal and spatial scales. Therefore, high spatiotemporal resolution detection of movement processes and movement-induced changes in brain structure and neural activity signals is an important technical foundation for conducting motor neuroscience research and has developed a set of tools based on traditional neuroscience methods combined with novel motor behavior decoding technologies, providing an innovative technical platform for motor neuroscience research. The protective effect of exercise in neurodegenerative diseases provides broad application prospects for its clinical translation. Applied research in motor neuroscience centers on deciphering the regulatory networks of neuroprotective molecules mediated by exercise. From the perspectives of exercise promoting neurogenesis and regeneration, enhancing synaptic plasticity, modulating neuronal functional activity, and remodeling the molecular homeostasis of the neuronal microenvironment, it aims to improve cognitive function and reduce the incidence of Parkinson’s disease and Alzheimer’s disease. This has also advanced research into the molecular regulatory networks mediating exercise-induced neuroprotection and facilitated the clinical application and promotion of exercise rehabilitation strategies. Multidimensional analysis of exercise-regulated neural plasticity is the theoretical basis for elucidating the brain-protective mechanisms mediated by exercise and developing intervention strategies for neurological diseases. Thus,real-time analysis of different neural signals during active exercise is needed to study the health effects of exercise throughout the entire life cycle and enhance lifelong sports awareness. Therefore, this article will systematically summarize the innovative technological developments in motor neuroscience research, review the mechanisms of neural plasticity that exercise utilizes to protect the brain, and explore the role of exercise in the prevention and treatment of major neurodegenerative diseases. This aims to provide new ideas for future theoretical innovations and clinical applications in the field of exercise-induced brain protection.

ObjectiveTo evaluate the therapeutic effect and mechanism of tendon-management and patella-movement therapeutic manipulation in the treatment of patellofemoral arthritis based on imaging evaluation. MethodsTotally 126 patients with patellofemoral arthritis were recruited and divided into a treatment group and a control group according to a randomised numerical table. The control group received routine sodium hyaluronate injection once a week for a total of 5 times； the treatment group received tendon-management and patella-movement therapeutic manipulation three times a week for four weeks. We compared the Western Ontario and McMaster University osteoarthritis index score （WOMAC）， visual analogue scale （VAS）， imaging indicators including patellar external displacement distance， patellofemoral fit angle， lateral patellofemoral angle， and patellofemoral index， and overall effectiveness evaluation between the two groups before and one week after treatment. ResultsThe total effective rate of the treatment group （45/54， 83.33%） was significantly higher than that of the control group （36/54， 66.67%，P＜0.05）. One week after the end of treatment， the VAS scores and WOMAC scores of both groups were lower than those before treatment in the same group （P＜0.01）， and the patellofemoral index and patellofemoral fit angle of the treatment group decreased compared with that of the control group （P＜0.05）. Compared with the pre-treatment， the distance of patellar external displacement， patellofemoral index， and patellofemoral fit angle decreased in the treatment group 1 week after the end of treatment， and the patellofemoral fit angle decreased in the control group （P＜0.05）. ConclusionThe therapeutic manipulation of tendon-management and patella-movement can correct the degree of patellar external displacement， alleviate joint pain symptoms， improve joint function， and achieve the goal of treating patellofemoral arthritis.

ObjectiveBased on functional near-infrared spectroscopy (fNIRS), we investigated the brain activity characteristics of gross motor tasks in children with autism spectrum disorder (ASD) and motor dysfunctions (MDs) to provide a theoretical basis for further understanding the mechanism of MDs in children with ASD and designing targeted intervention programs from a central perspective. MethodsAccording to the inclusion and exclusion criteria, 48 children with ASD accompanied by MDs were recruited into the ASD group and 40 children with typically developing (TD) into the TD group. The fNIRS device was used to collect the information of blood oxygen changes in the cortical motor-related brain regions during single-handed bag throwing and tiptoe walking, and the differences in brain activation and functional connectivity between the two groups of children were analyzed from the perspective of brain activation and functional connectivity. ResultsCompared to the TD group, in the object manipulative motor task (one-handed bag throwing), the ASD group showed significantly reduced activation in both left sensorimotor cortex (SMC) and right secondary visual cortex (V2) (P<0.05), whereas the right pre-motor and supplementary motor cortex (PMC&SMA) had significantly higher activation (P<0.01) and showed bilateral brain region activity; in terms of brain functional integration, there was a significant decrease in the strength of brain functional connectivity (P<0.05) and was mainly associated with dorsolateral prefrontal cortex (DLPFC) and V2. In the body stability motor task (tiptoe walking), the ASD group had significantly higher activation in motor-related brain regions such as the DLPFC, SMC, and PMC&SMA (P<0.05) and showed bilateral brain region activity; in terms of brain functional integration, the ASD group had lower strength of brain functional connectivity (P<0.05) and was mainly associated with PMC&SMA and V2. ConclusionChildren with ASD exhibit abnormal brain functional activity characteristics specific to different gross motor tasks in object manipulative and body stability, reflecting insufficient or excessive compensatory activation of local brain regions and impaired cross-regions integration, which may be a potential reason for the poorer gross motor performance of children with ASD, and meanwhile provides data support for further unraveling the mechanisms underlying the occurrence of MDs in the context of ASD and designing targeted intervention programs from a central perspective.

ObjectiveThe full name of vertebroplasty is percutaneous vertebroplasty (PVP). It is a clinical technique that injects bone cement into the diseased vertebral body to achieve strengthening of the vertebra. The research on the safety and efficacy of bone cement is the basis for clinical application. In this study, vertebroplasty is used to evaluate and compare the safety and efficacy of Tecres and radiopaque bone cement in experimental pigs, and to determine the puncture method suitable for pigs and the pre-clinical evaluation method for the safety and efficacy of bone cement. MethodsTwenty-four experimental pigs (with a body weight of 60-80 kg) were randomly divided into an experimental group (Group A) and a control group (Group B). Group A was the Tecres bone cement group, and Group B was the radiopaque bone cement group, with 12 pigs in each group. Under the monitoring of a C-arm X-ray machine, the materials were implanted into the 1st lumbar vertebra (L1) and 4th lumbar vertebra (L4) of the pigs via percutaneous puncture using the unilateral pedicle approach. The animals were euthanized at 4 weeks and 26 weeks after the operation, respectively. The L4 vertebrae were taken for compressive strength testing, and the L1 vertebrae were taken for hard tissue pathological examination to observe the inflammatory response, bone necrosis, and degree of osseointegration at the implantation site. ResultsThe test results of compressive strength between groups A and B showed no significant difference at 4 weeks and 26 weeks after bone cement implantation (P > 0.05). Observation under an optical microscope (×100) revealed that at 4 weeks postoperatively, both groups A and B showed that the bone cement was surrounded by proliferative fibrous tissue, with lymphocyte infiltration around it. The bone cement was combined with bone tissue, the trabecular arrangement was disordered, and osteoblasts and a small amount of osteoid were formed. At 26 weeks postoperatively, bone cement was visible in both groups A and B. The new bone tissue was mineralized, the trabeculae were fused, the trabecular structure was regular and dense with good continuity, and no obvious inflammatory reaction was observed. ConclusionIn experimental pig vertebrae, there were no significant differences observed in the compressive strength, inflammation response, bone destruction, and integration with the bone between Tecres and non-radiopaque bone cement. Both exhibited good biocompatibility and osteogenic properties. It indicates that using vertebroplasty to evaluate the safety and efficacy of bone cement in pigs is scientifically sound.

Objective: The study aimed to analyze the association between different types of obesity and dyslipidemia among rural primary and middle school students in Zhejiang Province, so as to inform strategies for prevention and control of childhood obesity and hyperlipidemia. Methods: As part of Nutrition Improvement Programme for Rural Compulsory Education Students, 1 244 participants were selected by stratified cluster random sampling in 5 counties of Zhejiang Province during September to December 2021. Physical examination, detection of blood lipid and questionnaire survey were conducted. The Chi -square test and Logistic regression analyses were used to assess the association between different types of obesity and dyslipidemia. Results: The prevalence rates of overweight, obesity, abdominal obesity, and hyperlipidemia were 15.11%, 12.46%, 17.60%, and 21.78%. Obesity and abdominal obesity were correlated to high risk of high triglycerides ( OR =3.97, 95% CI =2.54-6.20; OR =4.45, 95% CI =2.95- 6.72 )( P <0.05). Compared with the non overweight and obese group with normal waist circumference,the overweight and obesity group were correlated to high risk of high cholesterol ( OR=2.53, 95%CI =1.45-4.42, P <0.05). Abdominal overweight or obese group had the highest risk for dyslipidemia and triglycerides ( OR =1.82, 95% CI =1.33-2.48; OR =3.64, 95% CI =2.45-5.43) ( P < 0.05). Conclusions The prevalence rates of overweight, obesity, abdominal obesity, and hyperlipidemia are relatively high in rural primary and middle school students of Nutrition Improvement Programme for Rural Compulsory Education Students in Zhejiang Province. Abdominal obesity is a more important risk factor for hyperlipidemia. Waist circumference should be the focus of considerable attention.