緒論：探討透過內含加速規感測器的穿戴裝置在跑步機上進行不同速度的跑走時，裝置本身的信度與效度， 以及穿戴裝置所收取每分鐘前十大的加速度與走跑速度的相關性。方法：信度：一名健康男性受試者在左手 及左腳各放置一穿戴裝置，適應跑步機速度後，收取四種不同跑走速度資料，實驗為期四週，共收取十次資 料，以組內相關係數分析（Intraclass Correlation Coefficient，ICC）來檢測穿戴裝置的再測信度（Test-Retest Reliability）。效度：十名健康男性為受試對象，受試者在左手及左腳各放置一穿戴裝置和慣性感測器，適應跑 步機的速度後，收取不同速度下走跑的數據，分析兩者所收取的資料間是否相關。運動強度：探討穿戴裝置 測得數據與走跑速度是否相關。結果：穿戴裝置在上肢及下肢的信度考驗，以及穿戴裝置與效標之間的效度， 均有高度再測信度和高度相關（ICC=.962-.984，r = .883-.992）皆有高度相關；穿戴裝置測得數據也與走跑速 度有高度相關性（r = .869-.924）。結論：本實驗中穿戴裝置與儀器級慣性感測器做比較後擁有良好信效度，所 測得前十大加速度具有量化運動強度的可行性。

Introduction: The aim of this study was to assess the validity and the reliability of wearable sensor, which combined an accelerometer, in various speed of walking and running on a treadmill compare to laboratory Inertial Measurement Unit (IMU) sensor. In addition, this research also study about the relationship between body limb acceleration and moving speeds. Methods: Reliability: One healthy male subject was recruited and wore a wearable sensor on his left hand and a wearable sensor on his left leg. Four walking and running data in various speed were collected after the subject adapted to the speed on treadmill. The experiment lasts for four weeks and the data was collected ten times in total. Test-retest reliability was assessed using the Intra-class Correlation Coefficient (ICC). Validity: Ten healthy male subjects were recruited and wore a wearable sensor and an IMU sensor both on their left hands and left legs. After adapting to the speed on treadmill, researchers collected data of walking and running in various speed and compared the data between wearable sensors and IMU sensors. The Pearson correlation coefficient was conducted to examine the criterion-related validity. Exercise intensity: The Pearson product-moment correlation coefficient was conducted to examine the relationships between body limb acceleration and moving speeds. Results: Reliability: The test-retest reliability of left hand and left leg (ICC= .962-.984) were highly correlated. Validity: The validity results of wearable sensors compared with IMU sensors wore on left hands and legs (r= .883-.992) were highly correlated. Exercise intensity: In the entire walking-running period, body limb acceleration showed significant correlations with moving speeds (r = .869-.924). Conclusions: The wearable sensor has good validity and reliability compared to laboratory IMU sensor in this study. The measured acceleration has the feasibility of quantifying the exercise intensity.