Key result
An artificial neural network model using raw acceleration signals significantly reduced errors in estimating energy expenditure compared to IDEEA (P<0.01) and ActiGraph (P<0.001) monitors.
Why the study?
Does an artificial neural network model using raw acceleration signals improve the estimation of energy expenditure compared to standard accelerometers in healthy adults?
Population
102 healthy adults
Comparison
Feed-forward/back-propagation artificial neural… vs ActiGraph monitor and IDEEA monitor
Design
Other
Follow-up
nearly 24 h
Authors
Loading...
Supports more accurate EE estimation in healthy adults; leaves open validation in clinical populations.
Observational (n=102)
Does an artificial neural network model using raw acceleration signals improve the estimation of energy expenditure compared to standard accelerometers in healthy adults?
p-value: p=<0.01
An artificial neural network using raw acceleration signals provides more accurate estimates of energy expenditure than traditional linear regression models or multi-sensor arrays.
Rothney et al. (2007) conducted an observational in Healthy adults (n=102). Artificial neural network (ANN) model using raw acceleration signals vs. IDEEA monitor and ActiGraph monitor was evaluated on Mean absolute errors in estimating energy expenditure (p=<0.01). An artificial neural network model using raw acceleration signals significantly reduced errors in estimating energy expenditure compared to IDEEA (P<0.01) and ActiGraph (P<0.001) monitors.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: