Scientific Study

Temperature Algorithm Accuracy Testing

PKSE Refinement for ambient temperature

VO2 estimate from GoX Studio vs. Cosmed, August 2016

Dr. Lanny Griffin, Biomed, CalPoly, Analysis and Model Development

Objective: The purpose of this study was to modify the PKSE to include running and develop a predictive model which includes an effect of temperature.


Methods: We used the pilot data we collected during the base phase and some option phase data. The subjects walked at a constant 1.112 m/s pace on a level treadmill for 5 minutes, then increased their pace to 1.56 m/s for 3 minutes, then increased their pace to 2.24 m/s for 3 minutes, then increased their pace to 2.68 m/s for 3 minutes, then a 5 minute cool down at 1.112 m/s. Several subjects walked at 1.112 m/s with either 20% body weight load or 40% body weight load. The heart rate, respiration rate, and VO2 were measured using Cosmed. We normalized the energy expenditure by the body mass. VO2 was converted to watts and normalized by body mass.


Results and Discussion: We tested the model using data from the pilot study. The results are tabulated in table 1. The model predictions on average are 10 percent for both males and females.

Table 1: PKSE predictions for metabolic cost for subjects


  • Black LinkedIn Icon


GoX Labs

3110 N. Central Ave. Suite 153 

Phoenix, Arizona 85012

©2021 by GoXLabs