The Claim
A wrist-worn three-axis accelerometer employing a two-stage artificial neural network model predicts energy expenditure with a lower root mean square error (RMSE = 0.76 METs) during controlled treadmill walking and running at 2–9 km/h in healthy young adults aged 18–30 compared to linear, logarithmic, cubic, or single-stage artificial neural network models.
What the research says
Supports is higher
Support is ahead, but a single strong opposing study can change this.
These are independent scores, not a percentage. Higher-grade studies count more, so a single strong opposing study can outweigh several weaker ones.
A wrist-worn device using a two-stage artificial neural network measures energy expenditure during walking and running on a treadmill more accurately than simpler mathematical models in healthy adults aged 18–30.
See the scientific wording
A wrist-worn three-axis accelerometer using a two-stage artificial neural network model that separates walking and running phases predicts energy expenditure with lower error (RMSE = 0.76 METs) than linear, logarithmic, cubic, or single-stage ANN models during controlled treadmill walking and running at speeds of 2–9 km/h in healthy young adults aged 18–30.
The wrist device detects how the arm moves differently when walking versus running, and uses a two-part computer system to recognize these movement patterns and match them to the exact amount of energy being used, making the estimate more precise than simpler math methods.
What the research says
1 studyThis study found that a smart wristwatch that can tell the difference between walking and running, and uses a fancy computer program, guesses how many calories you burn more accurately than simpler math formulas when you walk or run on a treadmill.
Score breakdown, mechanism chain, raw evidence, ideal studies needed & 1 supporting studies
Not medical advice. For informational purposes only. Always consult a qualified healthcare professional before making health decisions.