Description

Four acceleration-based artificial neural networks were developed for estimating energy expenditure in adults using an ActiGraph worn at one of four wear locations- the right hip, right ankle, or either wrist. Adults performed simulated activities of daily living, locomotion, and other exercises and the model was cross-validated using leave-one-out cross validation. This paper also developed identical algorithms using only structured or only simulated free-living data for training but they are not currently included in the repository.

The following models are available. Root Mean Square Error (RMSE) is also provided.

• Artificial neural network for left wrist monitor (1.07 METs)

• Artificial neural network for right wrist monitor (1.09 METs)

• Artificial neural network for right ankle monitor (0.89 METs)

• Artificial neural network for right hip monitor (1.02 METs)

Phase Designation

(What's this?)

This model is in Phase 0 and 2.

Development/Validation

Sample: 24 (12 males) healthy adults, 18-80 years of age

Setting: Laboratory

Activities: Activities of daily living, cycling, walking and running, resistance exercise, stairs, stationary behaviors

Criterion: Oxycon Mobile (VO₂)

Accelerometer(s): ActiGraph GT9X on right hip, ankle, and both wrists

Validation approach: Leave-one-out cross-validation

Instructions

The R code can be used to read in a csv file with the necessary features and then use the neural network to predict energy expenditure for each 30-s window. There is a separate file for each of the four models. There is an example data sheet for each of the monitor wear locations (ankle, hip, left and right wrist) with the variables described in the data dictionary that need to be present in the csv file. Further instructions are in the R code. More information about R is found here.

This code relies on R software which can be downloaded for free at https://www.r-project.org/

Attached Files

Download all files as a .zip or download individual files below.

Code

Example left wrist data

Example right wrist data

Example hip data

Example ankle data

Data dictionary

Left wrist model

Right wrist model

Hip model

Ankle model

Note: Compared to the original source, code and sample data were edited for consistency, data dictionary was added, models were unchanged

Source Information

Reference:

Montoye, A. H., Conger, S. A., Connolly, C. P., Imboden, M. T., Nelson, M. B., Bock, J. M., & Kaminsky, L. A. (2017). Validation of accelerometer-based energy expenditure prediction models in structured and simulated free-living settings. Measurement in Physical Education and Exercise Science, 21(4), 223-234. https://doi.org/10.1080/1091367X.2017.1337638 Link to paper

Original source for code:

https://drive.google.com/file/d/0B-BgdTzyd2OxUDhwRWR6OTJwZmM/edit

Corresponding author: Alexander H.K. Montoye, montoyeah@alma.edu

Contact

Kimberly Clevenger at accelerometerrepository@gmail.com