at present, neural networks are becoming more and more complex, from several layers to dozens of layers or even more than 100 layers. The main advantage of deep network is that it can express very complex functions. It can learn features from different levels of abstraction, such as edge features at lower levels and complex features at higher levels.However, the use of deep networks is not always effective, because there is a very big obstacle - the disappearance of gradients: in very deep networks, gradient signals tend to approach zero very quickly, which makes the gradient descent process extremely slow.Specifically, in the process of gradient descent, the weight matrix product operation must be carried out in every step of back propagation from the last layer to the first layer, so that the gradient will drop exponentially to 0.(in rare cases, there is the problem of gradient explosion, that is, the gradient grows exponentially to the overflow in the process of propagation). Therefore, in the process of training, it will be found that with the increase of the number of layers, the rate of gradient decrease increases.Therefore, by deepening the network, although it can express any complex function, but in fact, with the increase of network layers, we are more and more difficult to train the network, until the proposal of residual network, which makes it possible to train deeper network[1].

Purpose Suboptimal health status (SHS) is a state between health and disease, has several adverse effects, although, its main underlying mechanism is still unclear. This study aimed to investigate SHS and its associated factors of adolescents. Methods A community-based cross-sectional study was conducted in the three different geographic locations of China (Shanxi, Guangzhou, and Tibet). A multidimensional sub-health questionnaire of adolescent (MSQA) is used to evaluate SHS. Independent two-sample K–S test was performed for the quantitative data as the non-parametric test, whereas Chi-square test method was applied to explore the difference of discrete variables data between groups. Then finally, multiple logistic regression analysis was applied to analyze the influential factors of SHS. Results Among 1461 respondents (between 15 and 18 years old), females proportion (56.47%) was higher than males (43.53%) where SHS was higher in Shanxi followed by Tibet and then Guangdong. The rural area, grade, lack of sleep time, home visit in a week, lack of exercise, a heavy burden of study, smoking, drinking, and fewer friends were the risk factors of SHS, while families living status, seeking help and extroversion were the protective factors. Conclusion SHS is significantly associated with behavior and lifestyle-related factors. For comprehensively prevention and control of the SHS, it is urgently needed to reduce the risk factors and enhance the protective factors among adolescents.

Suboptimal health status (SHS) is a state between health and disease, has several associated factors, although, its underlying mechanism is still unclear. This study aimed to investigate the status of SHS and its associated factors of high school students in three areas of China (Shanxi, Guangzhou, and Tibet). A multidimensional sub-health questionnaire of adolescent (MSQA) is used to evaluate SHS. Among 1461 respondents, females proportion 56.47% was higher than males 43.53% where SHS was higher in Shanxi followed by Tibet and then Guangzhou. The rural area, grade, lack of sleep, home visit in a week, lack of exercise, a heavy burden of study, smoking, drinking, and fewer friends were the risk factors of SHS, while, families living status, seeking help and extroversion were the protective factors. SHS is significantly associated with different influencing factors. For comprehensive prevention and control measures, reduce the risk factors and enhance the protective factors.