Unique Views on Obesity-Related Behaviors and Environments: Research Using Still and Video Images

Using Computer Vision to Annotate Video-Recoded Direct Observation of Physical Behavior

Direct observation is a ground-truth measure for physical behavior, but the high cost limits widespread use. The purpose of this study was to develop and test machine learning methods to recognize aspects of physical behavior and location from videos of human movement: Adults (N = 26, aged 18-59 y) were recorded in their natural environment for two, 2- to 3-h sessions. Trained research assistants annotated videos using commercially available software including the following taxonomies: (1) sedentary versus non-sedentary (two classes); (2) activity type (four classes: sedentary, walking, running, and mixed movement); and (3) activity intensity (four classes: sedentary, light, moderate, and vigorous). Four machine learning approaches were trained and evaluated for each taxonomy. Models were trained on 80% of the videos, validated on 10%, and final accuracy is reported on the remaining 10% of the videos not used in training. Overall accuracy was as follows: 87.4% for Taxonomy 1, 63.1% for Taxonomy 2, and 68.6% for Taxonomy 3. This study shows it is possible to use computer vision to annotate aspects of physical behavior, speeding up the time and reducing labor required for direct observation. Future research should test these machine learning models on larger, independent datasets and take advantage of analysis of video fragments, rather than individual still images.

The 2019 Conference on Health and Active Transportation: Research Needs and Opportunities

Active transportation (AT) is widely viewed as an important target for increasing participation in aerobic physical activity and improving health, while simultaneously addressing pollution and climate change through reductions in motor vehicular emissions. In recent years, progress in increasing AT has stalled in some countries and, furthermore, the coronavirus (COVID-19) pandemic has created new AT opportunities while also exposing the barriers and health inequities related to AT for some populations. This paper describes the results of the December 2019 Conference on Health and Active Transportation (CHAT) which brought together leaders from the transportation and health disciplines. Attendees charted a course for the future around three themes: Reflecting on Innovative Practices, Building Strategic Institutional Relationships, and Identifying Research Needs and Opportunities. This paper focuses on conclusions of the Research Needs and Opportunities theme. We present a conceptual model derived from the conference sessions that considers how economic and systems analysis, evaluation of emerging technologies and policies, efforts to address inclusivity, disparities and equity along with renewed attention to messaging and communication could contribute to overcoming barriers to development and use of AT infrastructure. Specific research gaps concerning these themes are presented. We further discuss the relevance of these themes considering the pandemic. Renewed efforts at research, dissemination and implementation are needed to achieve the potential health and environmental benefits of AT and to preserve positive changes associated with the pandemic while mitigating negative ones.

Measurement challenges for childhood obesity research within and between Latin America and the United States

Childhood obesity is a major public health challenge across Latin America and the United States. Addressing childhood obesity depends on valid, reliable, and culturally sensitive measurements. Such progress within and between countries of the Americas could be enhanced through better measurement across different age groups, different countries, and in sending and receiving communities. Additionally, better and more comparable measurements could accelerate cross-border collaboration and learning. Here, we present (1) frameworks that influenced our perspectives on childhood obesity and measurement needs across the Americas; (2) a summary of resources and guidance available concerning measurement and adaptation of measures for childhood obesity research; and (3) three major areas that present challenges and opportunities for measurement advances related to childhood obesity, including parental behavior, acculturation, and the potential to incorporate ethnographic methods to identify critical factors related to economics and globalization. Progress to reduce childhood obesity across the Americas could be accelerated by further transnational collaboration aimed at improving measurement for better surveillance, intervention development and evaluation, implementation research, and evaluation of natural experiments. Additionally, there is a need to improve training related to measurement and for improving access to valid and reliable measures in Spanish and other languages common in the Americas.

Field Test of a Passive Infrared Camera for Measuring Trail-Based Physical Activity

Introduction: Trails are ubiquitous and far-reaching, but research on the impact trails have on physical activity is limited by the lack of resource-efficient, accurate, and practical systematic observation tools. Commonly used infrared trail sensors count trail use and may broadly differentiate activity (i.e., bicyclist vs. pedestrian), but cannot detect nuances needed for outcomes research such as frequency, intensity, time, and type of activity. Motion-activated passive infrared cameras (PICs), used in ecological research and visitor management in wildlife areas, have potential applicability as a systematic observation data collection tool. Materials and Methods: We conducted a 7-month field test of a PIC as a systematic observation data collection tool on a hiking trail, using photos to identify each trail user's physical activity type, age, sex, and other characteristics. We also tallied hourly trail use counts from the photos, using Bland-Altman plots, paired t-tests, Concordance Correlation Coefficient, Kendall's Tau-b, and a novel inter-counter reliability measure to test concordance against concurrent hourly counts from an infrared sensor. Results: The field test proved informative, providing photos of 2,447 human users of the trail over 4,974 h of data collection. Nearly all of the users were walkers (94.0%) and most were male (69.2%). More of the males used the trail alone (44.8%) than did females (29.8%). Concordance was strong between instruments (p < 0.01), though biased (p < 0.01). Inter-counter reliability was 91.1% during the field study, but only 36.2% when excluding the hours with no detectable trail use on either device. Bland-Altman plots highlighted the tendency for the infrared sensor to provide higher counts, especially for the subsample of hours that had counts >0 on either device (14.0%; 694 h). Discussion: The study's findings highlight the benefits of using PICs to track trail user characteristics despite the needs to further refine best practices for image coding, camera location, and settings. More widespread field use is limited by the extensive amount of time required to code photos and the need to validate the PICs as a trail use counter. The future potential of PICs as a trail-specific PA research and management tool is discussed.