使用情境感知設計與實作一個脊髓損傷者的健康服務APP

Abstract

據中華民國脊髓損傷協會統計，每年約有一千多人因意外傷害或疾病造成脊髓損傷，其中有92%的傷者需要依靠輪椅來行動，且大部分傷者上肢仍具功能，須透過復健運動來強化其上肢功能，同時，也須避免上肢過度使用手推輪椅而造成二次傷害。近年來智慧行動裝置盛行，脊髓損傷者在生活中也幾乎人手一機，國外已有一些用來協助脊髓損傷者改善其生活的APP被提出。然而，目前這些APP大多僅只於提供單向衛教資料 (例如：復健運動影片或是相關醫療資訊等)，缺乏互動性社群與APP使用管理的智慧功能。對於脊髓損傷者而言，頻繁點擊的APP將會造成慢性傷害，需要減少使用時的點擊次數，降低負擔。但由於每個人的使用習慣各異，因此本論文開發一個智慧協助系統—SmartChair APP，以行動情境感知科技幫助解決這個客製化介面問題。SmartChair APP中透過行動裝置的感應器，可蒐集不同的情境資訊，幫助系統瞭解使用者需求即時提供適當服務。本論文使用一個具協同過濾機制的智慧情境感知推薦引擎—稱為iCASE (intelligence Context Awareness-based Suggestion Engine) ，對使用者行為進行資料的蒐集、分析與回饋。iCASE以xAPI記錄傷者的使用習慣，並加入情境感知技術蒐集在不同情境下的使用狀況，進行分析回饋以提供SmartChair APP動態最適介面佈局，達到更好的使用體驗。而蒐集到的數據與分析結果也可提供給親友家人、醫療團隊、相關軟硬體開發商，作為未來參考。

According to the Spinal Cord Injury Association of R.O.C, for each year, over 1,000 people are in spinal cord injury due to accidental injury or illness resulting. 92 percent of the injured need to use a wheelchair in everyday life. And, most of the injured still have the upper limb function. They need to avoid excessive use of the manual wheelchair, which might cause secondary damage of the upper extremity. Although some APPs were proposed to improve the quality of life for people with spinal cord injury, most of them provide only the health information, such as rehabilitation videos. They lack interactive community features. As everyone has his/her own unique habits, a customized interface should be provided to reduce the number of clicks and to lighten the injured for the burden of using APP. According to the above purpose, this thesis develops an intelligent assistance system, called the SmartChair APP, which bases on the context-awareness of mobile devices. Through this device, we can collect information for different situations to provide appropriate services. In this paper, we use a collaborative filtering mechanism with the wisdom of context-aware recommendation engine, called iCASE (intelligence Context Awareness-based Suggestion Engine). The iCASE will adopt xAPI for storing records of users' habits, and then we add context-awareness information for analysis. Finally, we can dynamically determine the optimal interface layout, providing the better user experience. The collected data and analysis can be applied for patients’ relatives, medical team, and even the medical-related software/hardware developers in the future.