Duration, frequency, and types of children's activities: Potential of a classification procedure

Transfer Learning Improves Accelerometer-Based Child Activity Recognition via Subject-Independent Adult-Domain Adaption

Wearable activity recognition can collate the type, intensity, and duration of each childs physical activity profile, which is important for exploring underlying adolescent health mechanisms. Traditional machine-learning-based approaches require large labeled data sets; however, child activity data sets are typically small and insufficient. Thus, we proposed a transfer learning approach that adapts adult-domain data to train a high-fidelity, subject-independent model for child activity recognition. Twenty children and twenty adults wore an accelerometer wristband while performing walking, running, sitting, and rope skipping activities. Activity classification accuracy was determined via the traditional machine learning approach without transfer learning and with the proposed subject-independent transfer learning approach. Results showed that transfer learning increased classification accuracy to 91.4% as compared to 80.6% without transfer learning. These results suggest that subject-independent transfer learning can improve accuracy and potentially reduce the size of the required child data sets to enable physical activity monitoring systems to be adopted more widely, quickly, and economically for children and provide deeper insights into injury prevention and health promotion strategies.

Changes in Body Composition and Anthropomorphic Measurements in Children Participating in Swimming and Non-Swimming Activities

BACKGROUND

Physical activity is a well-known means of obesity prevention, but the relationship between exercise frequency and body composition in children has not been thoroughly investigated.

OBJECTIVE

The aim of this study was to compare the body composition of children aged 11-12 who regularly performed swimming and other sports as an organized extra-curricular physical activity for a 12-week period.

METHODS

The study included 46 students who attended swimming classes and 42 students who participated in training activities in other sports, including, but not limited to, football, basketball and athletics. Body height and body composition were measured using a Tanita BC 418 MA analyzer. The students individually reported their rate of perceived exertion during training using the Pictorial Children's Effort Rating Table PCERT scale.

RESULTS

The weekly volume of training was substantially higher in the group of swimmers than in that playing other sports (12.3 h/week vs. 5.2 h/week, p < 0.01). After 12 weeks of training, body height and weight significantly increased in both groups (p < 0.001). However, the BMI value and adipose tissue content only increased in the group of non-swimmers. Swimmers perceived greater exertion during training than non-swimmers (7.1 vs. 5.8 on the PCERT scale, p < 0.01).

CONCLUSIONS

In early pubescent children, engaging in vigorous exercise such as swimming for at least 10 h a week may restrain the growth of adipose tissue. However, the variety of exercises that are typical of team sports, if performed for no more than 5 h a week, may be insufficient to restrain adipose tissue growth.

A Study on the Relationship between Urban Residents’ Perception of Recreational Sports and Their Participation in Recreational Sports: Based on Gender Differences

Background: The health of urban residents plays an important role in increasing the well-being of citizens, and therefore promoting the sustainable development of cities. Studies in Western countries have provided some evidence of the relationship between the urban residents’ perception of the recreational sports (RS), including perception of industry, space, experience, type, and support affecting recreational behaviors. Yet, evidence in China is not sufficient and, particularly, gender difference has not been discussed anywhere. Objectives: The present study aimed to see urban residents’ perception of RS and its effect on recreational sport participation by the residents. Method: The study was conducted in three big cities (Hangzhou, Chengdu, and Shanghai) of China to test the association between the urban residents’ perception of RS and their RS participation, in which gender differences were discussed. After taking their consent, a total of 764 participants (397 male and 367 female) aged from 15 to 65 were approached between April and October, 2018. The participants were given a survey to assess their perception and behavior of RS. T-test and the structural equation modeling were used to analyze gender differences in association between perception and behavior of RS. Results: For females, industry perception (β = 0.17, p = 0.00), space perception (females: β = 0.16, p = 0.00), and support perception (females: β = 0.09, p = 0.03) had a significant positive effect on RS participation. For males, industry perception (males: β = 0.13, p = 0.01) and type perception (males: β = 0.23, p = 0.00) showed a significant positive impact on participation of RS. However, space perception of males (β = −0.12; p = 0.01) and type perception of females (β = −0.11; p = 0.01) are negatively associated with their RS participation. The perception of experience was the most important factor influencing RS behaviors. Significant gender differences have appeared in all five different perception factors. The industry perception has significant and positive effects on both males and females’ RS participation. The type perception has significant positive effects on the male but negative on the female RS, while the space perception is just the opposite. The support perception has significant positive effects on the female RS practice but the influence is not obvious in the case of males. Conclusion: By adopting some measures, perception of RS may improve urban residents’ participation in RS.

Parents' Social Status and Children's Daily Physical Activity: The Role of Familial Socialization and Support

Physical activity is a health relevant factor, particularly in affluent societies where overweight and obesity are increasingly prevalent, even among children. Understanding the development of physical activity patterns in childhood is thus an important issue for health promotion. Following socialization theory, this study describes and explains differences in objectively measured moderate-to-vigorous physical activity (MVPA) in a socially and ethnically mixed sample of 6- to 11-year-old children in Germany. MVPA levels were objectively measured with accelerometers over the course of six consecutive days (Wednesday to Monday). Parents' attitudes and practices as well as the family's socio-economic status (SES) were assessed from the parents via questionnaires. Results indicate that MVPA levels of children vary with gender, but not with age and ethnicity. Moreover, parental SES, parental support for the child's sports activities, parents' own sport activities and the parents' belief in sports' capacities to foster personality development, character building and social integration significantly predict the MVPA level of children. It is concluded that interventions to promote MVPA among children need to take family interactions and lifestyles into account and should address families in socio-economically underprivileged areas.

Activity Recognition in Youth Using Single Accelerometer Placed at Wrist or Ankle

PURPOSE

State-of-the-art methods for recognizing human activity using raw data from body-worn accelerometers have primarily been validated with data collected from adults. This study applies a previously available method for activity classification using wrist or ankle accelerometer to data sets collected from both adults and youth.

METHODS

An algorithm for detecting activity from wrist-worn accelerometers, originally developed using data from 33 adults, is tested on a data set of 20 youth (age, 13 ± 1.3 yr). The algorithm is also extended by adding new features required to improve performance on the youth data set. Subsequent tests on both the adult and youth data were performed using crossed tests (training on one group and testing on the other) and leave-one-subject-out cross-validation.

RESULTS

The new feature set improved overall recognition using wrist data by 2.3% for adults and 5.1% for youth. Leave-one-subject-out cross-validation accuracy performance was 87.0% (wrist) and 94.8% (ankle) for adults, and 91.0% (wrist) and 92.4% (ankle) for youth. Merging the two data sets, overall accuracy was 88.5% (wrist) and 91.6% (ankle).

CONCLUSIONS

Previously available methodological approaches for activity classification in adults can be extended to youth data. Including youth data in the training phase and using features designed to capture information on the activity fragmentation of young participants allows a better fit of the methodological framework to the characteristics of activity in youth, improving its overall performance. The proposed algorithm differentiates ambulation from sedentary activities that involve gesturing in wrist data, such as that being collected in large surveillance studies.