Statistical methods to model and evaluate physical activity programs, using step counts: A systematic review

Using Computer Vision to Track Facial Color Changes and Predict Heart Rate

The current technological advances have pushed the quantification of exercise intensity to new era of physical exercise sciences. Monitoring physical exercise is essential in the process of planning, applying, and controlling loads for performance optimization and health. A lot of research studies applied various statistical approaches to estimate various physiological indices, to our knowledge, no studies found to investigate the relationship of facial color changes and increased exercise intensity. The aim of this study was to develop a non-contact method based on computer vision to determine the heart rate and, ultimately, the exercise intensity. The method was based on analyzing facial color changes during exercise by using RGB, HSV, YCbCr, Lab, and YUV color models. Nine university students participated in the study (mean age = 26.88 ± 6.01 years, mean weight = 72.56 ± 14.27 kg, mean height = 172.88 ± 12.04 cm, six males and three females, and all white Caucasian). The data analyses were carried out separately for each participant (personalized model) as well as all the participants at a time (universal model). The multiple auto regressions, and a multiple polynomial regression model were designed to predict maximum heart rate percentage (maxHR%) from each color models. The results were analyzed and evaluated using Root Mean Square Error (RMSE), F-values, and R-square. The multiple polynomial regression using all participants exhibits the best accuracy with RMSE of 6.75 (R-square = 0.78). Exercise prescription and monitoring can benefit from the use of these methods, for example, to optimize the process of online monitoring, without having the need to use any other instrumentation.

Leisure but Not Occupational Physical Activity and Sedentary Behavior Associated With Better Health

OBJECTIVE

This study explores relations between occupational and leisure-time physical activity (OPA, LTPA) and sedentary behavior (OSB, LTSB) and several health outcomes.

METHODS

A total 114 full-time workers had their body composition, waist circumference, height, weight, resting heart rate, and resting blood pressure measured. ActivPal monitor measured physical activity behaviors. Stress, mood, and pain were measured with ecological momentary assessment. General linear models were used to examine the relationship between high and low OPA, LTPA, OSB, and LTSB with each health outcome while controlling for covariates.

RESULTS

The high LTPA group had lower body mass index (BMI) (P = 0.04) and better mood (P = 0.007) than the low LTPA group. The high LTSB group had higher systolic blood pressure (P = 0.001), higher diastolic blood pressure (P = 0.01), higher BMI (P = 0.027), higher body fat percentage (P = 0.003), higher waist circumference (P = 0.01), and worse mood (P = 0.032) than the low LTSB group. No differences were found between OPA and OSB groups.

CONCLUSIONS

These findings suggest there may be differential relations between PA and SB accumulated during leisure versus occupational time.

Gamified Text Messaging Contingent on Device-Measured Steps: Randomized Feasibility Study of a Physical Activity Intervention for Cancer Survivors

BACKGROUND

Physical activity can confer diverse benefits on cancer survivors. Unfortunately, many cancer survivors are not sufficiently active. The efficacy of physical activity interventions for this population may be increased by grounding them in Self-Determination Theory (SDT). Combining game design elements with wearable technologies may be a useful and scalable approach to targeting SDT constructs to promote cancer survivors' physical activity.

OBJECTIVE

The primary aim of this study is to evaluate the feasibility and acceptability of Steps2Health, a physical activity intervention for cancer survivors. It also aims to investigate the effects of the intervention on motivation, physical activity, and step count.

METHODS

We randomized 78 insufficiently active cancer survivors to an experimental or comparison group. All participants received a physical activity tracker. The experimental group participants also received a set sequence of multimedia messaging service messages that were triggered in real time by meeting predetermined cumulative step count totals. Messages presented information about a virtual journey and included photographs and vivid descriptions of locations to increase autonomous motivation. Additional messages targeted perceptions of relatedness (eg, role modeling) and competence (eg, facilitating mastery experiences). We administered pre- and postintervention surveys and conducted 15 individual interviews to evaluate the intervention. We performed directed content analysis of qualitative data and conducted mixed effects linear modeling to investigate participants' changes in motivation, self-reported physical activity, and device-measured step counts.

RESULTS

There was minimal loss to follow-up (3/78, 4%), the device wear rate was high (2548/3044, 83.71% of days), and technical problems with messaging based on real-time step counts were limited. Our qualitative data analysis revealed 3 overarching themes: accessibility, autonomous motivation, and relatedness. Participants successfully navigated the technological aspects and game design elements of the intervention. Participants found messages targeting autonomous motivation and competence or self-efficacy to be enjoyable and compelling, but one feasibility criterion for participant engagement (response rate to text messages) was not met. Messages targeting relatedness were less highly rated than the messages targeting autonomous motivation and competence or self-efficacy. During the intervention, both groups increased their motivation for physical activity (B=0.16; 95% CI 0.01 to 0.30; P=.04; d=0.49), and assignment to the experimental group was associated with increased self-reported leisure activity score (B=10.78; 95% CI 3.54 to 18.02; P=.005; d=0.64). The experimental group had greater increases in daily step counts over time (B=322.08; 95% CI 54.01 to 590.15; P=.02; d=0.28).

CONCLUSIONS

This study supports the feasibility of using real-time game design elements to target SDT constructs and increase cancer survivors' physical activity. Overall, our findings support the acceptability of the Steps2Health intervention, but fostering active participant engagement and targeting relatedness may present additional challenges. Steps2Health may help cancer survivors increase their physical activity levels.