近年來,隨著醫療科技已開發國家皆漸漸進入高齡化社會,跌倒是造成台灣地區高齡者因事故死亡的第二大原因,然而造成高齡者跌倒的最主要因素為平衡控制能力退化,因此若能有好的平衡控制訓練方式,除了能提升個人的平衡能力外還能夠預防跌倒發生所帶來的嚴重後果。過去已有應用虛擬實境(Virtual reality;VR)與體感遊戲在輔助訓練平衡之研究並針對單腳站立人形框參數設計做探討,然而過去使用Animazoo動作捕捉器設備其價格過於昂貴一般人難以取得且需要耗費一段時間在做軟體設定工作;再者,評估方法,僅蒐集人形框與人模碰撞次數做為平衡控制數據,此方式無法表示受測者在不同人形框參數下之身體搖晃程度,另外實驗之受測者皆為年輕人無考慮不同平衡控制退化高齡者。 本研究以Kinect體感捕捉器為基礎發展平衡訓練體感遊戲場景,目的比較在不同訓練場景下,健康年輕人與高齡者進行平衡控制訓練時有何影響,以及不同平衡控制退化程度之高齡受測者在增加訓練次數後,平衡控制表現趨勢,本實驗設計四種參數,人形框大小、人形框框移動時間為1秒或2秒、抬腳角度為45度與90度與抬腳速度為1秒與2秒共八種不同的訓練場景,實驗時受測者站在測力板上觀看前方螢幕,Kinect驅動虛擬場景中的虛擬人模,將人模維持於人形框內,並探討在不同的參數設計下受測者COP移動軌跡與碰撞次數差異。 實驗結果顯示人形框四種參數對於年輕受測者的碰撞有顯著影響,而高齡者僅移動時間、抬腳角度與抬腳速度有影響。以COP移動軌跡結果顯示,年輕受測者之抬腳速度對於MDIST-ML與TOTEX、移動時間對於MDIST-AP有顯著影響;高齡受測者之抬腳速度對於TOTEX、移動時間對於MDIST-AP、MDIST-ML與AREA-SW有顯著影響,其餘則無;再者,高齡受測者在經由五次的訓練後無顯著的改善平衡控制表現。本研究技術可讓未來針對高齡者設計平衡控制訓練體感遊戲中參數設計做考量,以調整平衡控制能力訓練的困難度。
Fall prevention has become a more and more important issue as fall-down has been cited as the number one reason to cause accidental death of aged adults. Static balance training is an effective method to increase the balancing capacity of elderly and reduce their fall risks. This paper proposes combined virtual reality (VR) with Kinect system to practice balance control training. Our method lets the subjects standing under the force plate and watch the screen in a virtual environment to follow a 3D posture frame(PF) do stand on one foot . A usability evaluation experiment is conducted and the results show that the parameters of PF design will affect the subjects posture control performance. Our approach is superior to existed methods in several aspects. First, the training process could be customized by modifying the parameters of the PF. Secondly, the training process is accurate because the trainee has to follow the PM. Thirdly, the training process could be quantitatively evaluated by detecting and analyzing the collision information and Center of point(COP). The results show that the visual stimulations created by different PF designs could affect the trainee’s posture control ability and quality.