Ideal Combinations of Acceleration-Based Intensity Metrics and Sensor Positions to Monitor Exercise Intensity under Different Types of Sports

Abstract

This study quantified the strength of the relationship between the percentage of heart rate reserve (%HRR) and two acceleration-based intensity metrics (AIMs) at three sensor-positions during three sport types (running, basketball, and badminton) under three intensity conditions (locomotion speeds). Fourteen participants (age: 24.9 ± 2.4 years) wore a chest strap HR monitor and placed three accelerometers at the left wrist (non-dominant), trunk, and right shank, respectively. The %HRR and two different AIMs (Player Load per minute [PL/min] and mean amplitude deviation [MAD]) during exercise were calculated. During running, both AIMs at the shank and PL at the wrist had strong correlations (r = 0.777-0.778) with %HRR; while other combinations were negligible to moderate (r = 0.065-0.451). For basketball, both AIMs at the shank had stronger correlations (r = 0.604-0.628) with %HRR than at wrist (r = 0.536-0.603) and trunk (r = 0.403-0.463) with %HRR. During badminton exercise, both AIMs at shank had stronger correlations (r = 0.782-0.793) with %HRR than those at wrist (r = 0.587-0.621) and MAD at trunk (r = 0.608) and trunk (r = 0.314). Wearing the sensor on the shank is an ideal position for both AIMs to monitor external intensity in running, basketball, and badminton, while the wrist and using PL-derived AIM seems to be the second ideal combination.

Keywords: acceleration; racquet sports; running; team sports; wearable electronic devices.

Figure 1

Accelerometers were placed on the left wrist (a), trunk (b) and shank (c).

Figure 2

Custom basketball exercise routine (a) and badminton six-point footwork (b).

Figure 3

Change of %HRR among three exercise intensity levels under different sports (n = 14). Bonferroni post hoc test: significantly different than ^a Level 1, ^b Level 2, and ^c Level 3 (p < 0.05); Wilcoxon signed rank test: significantly different than * Level 1, ^† Level 2, and ^‡ Level 3 (p < 0.05).

Figure 4

Change of PL/min among three exercise intensity levels under different sports (n = 14). Bonferroni post hoc test: significantly different than ^a Level 1, ^b Level 2, and ^c Level 3 (p < 0.05); Wilcoxon signed rank test: significantly different than * Level 1, ^† Level 2, and ^‡ Level 3 (p < 0.05).

Figure 5

Change of MAD among three exercise intensity levels under different sports (n = 14). Bonferroni post hoc test: significantly different than ^a Level 1, ^b Level 2, and ^c Level 3 (p < 0.05); Wilcoxon signed rank test: significantly different than * Level 1, ^† Level 2, and ^‡ Level 3 (p < 0.05).

Figure 6

The within-sport Pearson (r) (n = 42) and across-sports Spearman (ρ) (n = 126) correlation coefficient for %HRR and PL.

Figure 7

The within-sport Pearson (r) (n = 42) and across-sports Spearman (ρ) (n = 126) correlation coefficient for %HRR and MAD.