目的:本研究分析太極拳單招的力學特性,搭配太極拳師父的單招教學錄影與體感偵測技術,開發一套擴增實境太極拳訓練系統,並招募高齡者參與試驗,驗證其對改善平衡能力的成效。方法:太極拳選用楊氏太極24式,以各單招的雙腳站姿與執行時身體重心移動軌跡長度來區分執行難度,再依據高齡者自主移動重心能力的表現挑選適合的訓練單招。擴增實境功能使用Kinect for Windows進行動作偵測,將使用者執行訓練時的即時影像與偵測的骨架資訊疊合顯示,與太極拳師傅執行同一單招的資訊比對,立即提供動作提示。系統開發完成後招募高齡者參與試驗,為期8週,每週3次、每次30分鐘。使用伯格式量表(Berg Balance Scale, BBS)、計時起走測試(Timed Up and Go Test, TUG)、功能性前伸測試(Functional Reach Test, FR)與下肢肌肉力量測試,來評估介入前後的表現差異,統計顯著差異水準訂在p<.05。結果:本研究共招募高齡者8人(平均年齡為70.0±4.2歲),訓練後,BBS平均進步了3.5±1.0分(p=.011),TUG平均快了2.0±0.4秒(p=.011),FR平均增加了4.3±2.4公分(p=.012),下肢肌力測試平均進步了3.8±0.9公斤,皆達顯著差異。結論:本研究開發了一套擴增實境太極拳訓練系統,參與訓練的高齡者經過8週的訓練後,平衡功能及下肢肌力表現有顯著提升。結合擴增實境進行太極拳訓練應對改善高齡者平衡表現有良好助益。
Purposes: The study analyzes the mechanical characteristics of Tai-Chi's single strokes and uses Tai- Chi's single stroke teaching videos and body motion detection technology to develop an augmented reality (AR)-assisted coaching system, and recruit elderly to participate in the experiment to verify the training effects on balance. Methods: The Yang's Tai-Chi 24 styles were used and the difficulty of each style was graded by the standing posture on both feet and needed center of pressure (COP) movement trajectory length. The suitable single styles selected for training were according to the subject's ability to autonomously move the COP. Augmented reality function used Kinect for Windows to detect the subject's movement and displayed the subject's real-time image and skeleton information simultaneously. The subject's skeleton information was compared with the information of the Tai-Chi master performing the same single style to provide action prompts immediately. Senior subjects were recruited to verify the effect of the augmented reality-assisted training on balance ability. The duration lasted 8 weeks, 3 times a week, and 30 minutes each section. The Berg Balance Scale (BBS), Timed Up and Go Test (TUG), Functional Reach Test (FRT), and lower limb muscle strength tests were used to assess pre-intervention and post-intervention difference, the statistically significant level was set at p < .05. Results: Eight seniors were recruited (age: 70.0 ± 4.2 years old). After intervention, BBS was improved by an average of 3.5 ± 1.0 points (p = .011), and TUG was decreased by an average of 2.0 ± 0.4 seconds (p = .011) and FRT was increased by an average of 4.3 ± 2.4 cm (p = .012). In the muscle strength test, the lower limb muscle group improved by an average of 3.8 ± 3.6 kg, with significant differences in all 16 muscle groups. Conclusion: The improvement can be observed in clinical balance tests and lower limb muscle strength on elderly subjects with the AR-assisted Tai-Chi coaching system developed in the study with 8-weeks training. The AR-assisted Tai- Chi coaching system should help improve the balance performance of the elderly.