足壓量測資料(Plantar Pressure Data)在運動復健領域中為重要判斷依據,得以診斷足底症狀,其研究數據包含足弓指標、左右腳足壓分區、足壓六區分區等類型共20種參數,造成多維度複雜性以及異常個案資料,造成資料觀察不易性,所以本研究建置資料庫平台,將足壓量測資料整理後,提出離群值(Outlier)辨識方法找尋離群值資料,提出差異可視化方法使運動復健領域學者更易觀察群組資料差異性。本研究資料主要針對各大專院校專業運動員樣本,共1006筆資料,分26類運動項目,用來研究各運動項目衍生的足部職業病症,來源取自於運動復健領域學者提供。本論文將足壓資料群組化後,利用平行座標(Parallel Coordinate)顯示方法、足底六區直方圖顯示方法、散點圖顯示方法去實作差異可視化(Comparative Visualization)概念,進行比對群組間各參數平均值以及標準差範圍或觀察其分佈行為,並利用差距標準差值辨識離群值(Outlier)資料,根據平行座標顯示方法協助運動復健領域學者進一步觀察離群值資料。並針對羽球運動群組、網球運動群組、桌球運動群組這三種相似運動行為的運動項目,利用本研究提出的方法進行分析研究。利用複數群組差異可視化,直接觀察這三種運動項目之間差異性,進而根據本研究顯示結果及自身專業判斷研究群組步態行為及判斷群組衍生足部症狀。利用本研究離群值辨識方法,能讓運動復健領域學者省去辨識時間,直接研究單一群組離群值資料。
Plantar pressure data is an important foundation for sports medicine. It includes 20 parameters from the index arch, left foot and right foot plantar pressure. It causes data observation inconvenient because of complicated property and outlier. We build database platform and import plantar pressure to database. Using outlier identification and comparative visualization to help sports medicine researchers studying plantar pressure data. Data source comes from sports medicine researcher who research on college athletes. The dataset contains total 1006 data collected from 26 sports. In order to research foot symptoms. This paper proposes methods to visualize plantar pressure data using comparative visualization. Parallel coordinate, plantar bar chart and scatterplot to compare average value and standard deviation of dataset. We search outliers using range standard deviation value and display them. It can help sports medicine researchers to observe outlier. We research similar exercise behavior sports like snooker, tennis, badminton. Sports medicine researchers can use comparative visualization to compare difference datasets. It can save sports medicine researchers time in operate plantar data and help them easy to observe difference between sports. Further, they can diagnose foot symptoms from clustering datasets. Sports medicine researchers can use outlier identification to find outlier. It saves sports medicine researchers time in identifying outlier.