Improving energy expenditure estimates from wearable devices: A machine learning approach

ABSTRACT

A means of quantifying continuous, free-living energy expenditure (EE) would advance the study of bioenergetics. The aim of this study was to apply a non-linear, machine learning algorithm (random forest) to predict minute level EE for a range of activities using acceleration, physiological signals (e.g., heart rate, body temperature, galvanic skin response), and participant characteristics (e.g., sex, age, height, weight, body composition) collected from wearable devices (Fitbit charge 2, Polar H7, SenseWear Armband Mini and Actigraph GT3-x) as potential inputs. By utilising a leave-one-out cross-validation approach in 59 subjects, we investigated the predictive accuracy in sedentary, ambulatory, household, and cycling activities compared to indirect calorimetry (Vyntus CPX). Over all activities, correlations of at least r = 0.85 were achieved by the models. Root mean squared error ranged from 1 to 1.37 METs and all overall models were statistically equivalent to the criterion measure. Significantly lower error was observed for Actigraph and Sensewear models, when compared to the manufacturer provided estimates of the Sensewear Armband (p < 0.05). A high degree of accuracy in EE estimation was achieved by applying non-linear models to wearable devices which may offer a means to capture the energy cost of free-living activities.     

Encouraging problem solving

Children seem to be natural problem solvers and delight in the challenges that are provided for them. Teachers who are careful observers of what children do can begin to provide many opportunities for helping them build their skills in problem solving. At the same time, it is important to let children create and solve some of their own newly discovered problems. A balance of both seems to be important to solving problems.Janis Bullock is Instructor of Early Childhood Education at Montana State University in Bozeman.     

Measures of cognitive structure: Do they assess learning at the level of comprehension?

One purpose of this study was to determine whether cognitive structure, assessed by psychometric measures of concept interrelatedness, can be developed when the students initially do not know what concept relationships exist and what they mean. The second purpose was to apply those measures to a learning situation that has produced a nonspecific transfer effect, i.e., the facilitative effect of concrete examples on learning abstract passages, to attempt to explain this effect more completely. Five groups of 20 students each read two prose passages and took recall and structure assessment tests on the second passage. Results (1) indicated that the nonspecific facilitative transfer effect was replicated and (2) offered some support for the contention that the cognitive structure which proximity measures assess can be trained to correspond to content structure, but that related recall remains low. With resolution of some of the methodological issues surrounding these measures, however, clearer explanation of transfer effects and assessment of higher order learning may be facilitated.     

Partnership for Faith-Based Leadership Development: An Educational Model

Parochial school leaders traditionally rise from the ranks of the teaching staff into the principal position, even if unprepared. Through a unique collaborative model, an outside partner and a university developed a program for formally preparing individuals to be Jewish day school leaders. The process of creating this model resulted in a reconceptualization of school leadership.     

The very model of a modern major (media) candidate: Colin Powell and the rhetoric of public opinion

In the surfacing stage of the 1995 Presidential campaign, media attention was focused on a non‐candidate, Retired General Colin Powell. This study examines the media's justification for such attention, in light of the absence of Powell's public instigation of a campaign. Coverage of Powell as a potential candidate reflects current media practice with regard to speculative reporting and the assertion of objectivity. Application of Burke's pentad to specific coverage in TIME and Newsweek illustrates how the media uses constructions of public opinion to rhetorically justify campaign coverage of a non‐candidate, transferring agency from the candidate to the “Public.”;     

Assessing working memory capacity in a non-native language

The present studies directly test the usefulness of two English-language working memory capacity (WMC) assessments with two samples of students whose native language was not English. Participants completed two widely used complex span tasks, Reading Span (RSpan) and Operation Span (OSpan), in English. To determine whether the well-established relationship between WMC and Raven's Advanced Progressive Matrices (RAPM) would be observed when span tasks were not given in the native language of the participant, span scores were regressed against performance on the RAPM. Results indicated that while OSpan was a reliable and valid predictor of RAPM in non-native-English speakers, RSpan administered in English was not.