Smartwatch-Derived VO2max Prediction Model for Korean Adults: Utilizing Heart Rate and GPS Data from the 12-Minute Cooper Test

Abstract

Background/Objectives: Recent technological advancements enable smartwatches to measure running distance and heart rate using wearable sensors. This study aimed to analyze the validity of the 12 min Cooper test using a smartwatch and to develop an accurate VO2max prediction model for Korean adults. Methods: A total of 104 adults (53 males: age 35.00 +/- 6.1 years, BMI 24.71 +/- 3.13; 51 females: age 34.82 +/- 6.07 years, BMI 22.24 +/- 2.66) participated. VO2max was measured using a maximal graded treadmill test. Participants performed the Cooper test while wearing a smartwatch, which collected average heart rate, peak heart rate, and running distance. Sex, height, and weight were also included as predictor variables. Multiple regression analysis was conducted to develop a VO2max prediction equation. Model accuracy was assessed using R2 and the standard error of the estimate (SEE). Results: The developed VO2max prediction equation was VO2max = 27.620 + 6.358 (sex; male = 1, female = 0) - 0.012 (height) - 0.202 (weight) - 0.036 (mean HR) + 0.039 (peak HR) + 0.012 (distance) (R2 = 0.853, SEE = 3.176 mL/kg/min, p < 0.001). The intra-class correlation coefficient (ICC) between measured and predicted VO2max using the smartwatch was 0.961, compared to 0.925 for traditional methods. The explanatory power was 86.0% (SEE = 3.024 mL/kg/min) versus 81.0% (SEE = 3.516 mL/kg/min). Conclusions: The smartwatch-based VO2max prediction model demonstrated higher accuracy than traditional methods. This equation is recommended for more precise VO2max estimation in Korean adults.