目的：探討RT3 三軸加速規與氣體分析儀在桌球正手擊球測試間之相關性，並檢驗不同送球頻率及不同配戴位置之差異。方法：20 名平均20.25 ± 1.65 歲健康男性桌球選手，過去半年無任何上肢傷害。參與者進行3 種不同送球頻率的桌球正手擊球測試，實驗中配戴RT3 於下背部、肩、肘和腕關節，並以氣體分析儀量測運動時之能量消耗。參與者隨機進行3 種送球頻率60、75、90 球/分鐘的正手擊球測試連續5 分鐘，各測試間隔5 分鐘以上。記錄5 分鐘內RT3 三軸向量與能量消耗預估值與氣體分析儀之耗氧量數值。統計方法以皮爾森相關係數分析不同儀器間參數之相關性，再以相依樣本二因子變異數分析分析RT3 與氣體分析儀不同送球頻率測試間與RT3 在不同位置之差異。結果：在3 種不同測試，肘、腕關節的RT3 所測得之參數與氣體分析儀之耗氧量呈現中度到高度的相關性 (r = 0.39 - 0.75, p ＜ .01)。在3 種不同送球頻率測試中，2 種儀器所測得的能量消耗及向量大小，會隨著送球頻率的增加而上升，且不同送球頻率間皆有顯著差異。RT3 在相同送球頻率測試中4 個位置的比較，所測得的數值依小至大順序為下背部、肘關節、肩關節、腕關節。在60 球/分鐘和75 球/分鐘的測試中，位置間皆有顯著差異 (p ＜ .05)，但在90 球/分鐘的測試中，肩關節與肘關節並無顯著差異，其餘位置間皆有顯著差異 (p ＜ .05)。結論：RT3 應用在桌球正手擊球測試中與氣體分析儀有中至高度的相關性，並可區辨不同送球頻率的測試，且以配戴於肘關節位置較佳。

Purpose: The purposes of this study were to investigate the correlation between RT3 triaxial accelerometer and indirect calorimetry measurements; to compare data from three different forehand table-tennis stroke frequencies and from RT3 worn at different anatomic locations. Methods: Twenty young healthy male table-tennis athletes (20.25 ± 1.65 years) with no known history of upper limbs injuries in the previous 6 months volunteered to participate in this study. The triaxial accelerometer (RT3, Stayhealthy, Inc.) and indirect calorimetry (SensorMedics, Vmax 29) were used simultaneously to measure the energy expenditure and oxygen consumption during exercise. Four accelerometers were attached at lower back (L3), dominant side of shoulder, elbow, and wrist. Each subject randomly adopted tree forehand table-tennis trials at frequency of 60 balls/min, 75 balls/min, 90 balls/min for 5 minutes. Pearson correlation coefficient was used to examine the correlation between the RT3 and indirect calorimetry measurements. Repeated measures two-way ANOVA was used to test any significant difference among three stoke frequencies and four RT3 locations. Results: Correlations between RT3 energy consumption and indirect calorimetry, RT3 vector magnitude and indirect calorimetry (for RT3 at elbow and wrist) are moderate to high (r = 0.39-0.75, p ＜ .01). All parameters were lower for 60 balls/min than 75 balls/min and 90 balls/min in all test conditions. For all test conditions, the vector magnitude and energy expenditure measured by RT3 were lowest at the low back than elbow, shoulder and wrist (low back ＜ elbow ＜ shoulder ＜ wrist). No statistically significant difference were found between RT3 at shoulder and elbow. Conclusion: We concluded that RT3 was suitable in distinguishing different forehand table-tennis stroke frequencies across all wearing locations and had moderate to high correlation with indirect calorimetry.