在這數十年間，中風一直是全球重要的疾病之一，尤其是步入高齡化社會的已開發國家，隨著醫療進步，老年人口漸多，中風問題也日漸受重視。本研究針對上肢中風復健進行各項指標分析，嘗試找出可供治療師進行評估參考的資訊。 研究中分為兩大部分，第一是找出對應到本系統的各項運動指標，並且分析運動指標對於臨床評估量表會有何影響；第二是運動指標建立後，以大量病人的數據做為群體樣本利用圖形識別的方法，如類神經網路與支持向量機進行分類。有別於臨床評估量表必須進行多個測試，才能得知病人的級別分類，我們可以直接透過遊戲的結果與系統所截取的運動數據進行分析即可得到病人的分類狀況。 本研究結果顯示，部分的運動指標確實對於評估成果有著相當程度的影響，並且病人在遊戲後對於運動指標的數值也有著明確的進步。在分類系統模型的建立，以現有的運動數據進行分類，確實也有著一定的辨識率，因此在建立分類系統模型是有一定的可行性的。 ;In the past decades, stroke has been one of the world′s major diseases. Especially in the developed countries and aging society, with medical advances, an increasing elderly population, stroke problems are increasingly valued. In this study, we analyzed the different motor index for the stroke rehabilitation with upper extremity, try to find a new indicators information for therapists. Study is divided into two parts, the first is to identify motor index that correspond to the movement of the system, and analyzed if the motor index would impact result of clinical assessment scale. The second, when found out the motor index, we used large number of patient′s data to classify groups with pattern recognition, like neural network or SVM. Unlike clinical assessment scale multiple tests must be carried out in order to know the patient′s level of classification, we can analyze the patient′s classification status can be obtained directly through the game′s outcome and the motor index of data with system. The results of this study shows that part of the motor index for assessment scale have a considerable degree of influence. And after rehabilitation, the numerical of motor index also has definite progress. We used motor data that collected in system to establishing classification model also has a certain recognition rate. Thus establishing the classification model is a certain degree of feasibility.
