整合穿戴式裝置於個人健康資訊系統之開發 - 以HRV量測及應用為例

因應慢性病高居國民十大死因、長期照護之人力不足等健康議題，同時考量現有技術與設備後發現穿戴式裝置在技術與普及率上已漸趨成熟；穿戴式裝置的特性非常適合輔助長期照護之進行，且相較於硬體的發展軟體仍處於萌芽階段，欲將此一技術導入使用者的生活需求中，資料處理與分析的部分仍須朝準確度的提高與分析服務的多樣性努力。因此，研究目的為將開發一健康管理系統，整合現有穿戴式裝置並藉由整合分析方法提高準確度、確保資料正確性等方式提高使用者感受。本研究以整合穿戴式裝置之健康管理系統為目標，在實行系統分析後分別規劃參數分析模組與資料傳輸防錯模組，量測端整合的穿戴式設備為可量測心率之運動手錶，參數分析模組在接收穿戴式裝置所量測的資料後進行頻域參數的分析，並依照不同的限制條件切換Fast Fourier Transform和Lomb-Scargle Periodogram分析方法；資料傳輸防錯模組將轉換後參數進行循環冗餘檢查碼和CRC-CCITT生成多項式編碼並傳送，資料接收端則以相同的生成多項式進行解碼動作並確認資料是否有誤，若發生錯誤則通知重新傳送；整合上述模組以實現可檢索自身健康資料的健康管理系統，並實施SUS使用者易用性評估以瞭解是否符合使用者需求與感受。本研究已具體實現整合穿戴式裝置之健康管理系統，簡易操作介面讓使用者可進行設備配對、資料與轉換後參數的即時呈現，針對資料傳輸防錯模組進行固定資料長度模擬隨機位置產生錯誤資料測試錯誤偵測率，測試結果得到將近100%的檢錯率，循環冗餘檢查碼同時具有高效率的特性可減少錯誤偵測產生的負擔，最後針對系統在使用上實施SUS易用性評估結果表示使用者大致上對於此系統的簡潔介面與操作流程大致上滿意，正向提問上關於系統的整合性滿意度相對較低，因此可再深入瞭解是否有使用者所期待或應具備之功能未整合至系統內。對於未來的可發展方向可評估在資料傳輸防錯模組內加入錯誤更正碼的效益與合適情境；對於資料的應用可結合進階之資料探勘方法已提供更加彈性與個人化的分析服務。

關鍵字

個人健康；穿戴式； HRV

並列摘要

In view of the health problems of chronic diseases, such as the top 10 causes of death, long-term care and the consideration of existing technologies and equipment, it is found that wearable techniques are becoming more mature and widespread. The characteristics of wearable devices are very suitable for long-term care. And the development of software compared to the hardware is still in the embryonic stage. For import this technology into the user's life needs, still need to improve the accuracy and service diversity of data processing and analysis. Therefore, the purpose of the study is to develop a health management system, integrate existing wearable devices and improve the accuracy of the data processing by integrating the analytical methods to ensure the correctness of the outcome. In this study, we focus on developing the health management system by integrating wearable device. After the system analysis, we plan the parameter analysis module and the data transmission error prevention module. The parameter analysis module analyzes the frequency domain parameters after receiving the data measured in the wearable device and switches the fast Fourier transform and the Lomb-Scargle Periodogram analysis method according to the different restriction conditions. The data transmission error prevention module converts the parameters into cyclic redundancy check code by the CRC-CCITT generation polynomial and transmitted. The data receiving end performs the decoding operation with the same generating polynomial and confirms whether the data is erroneous and informs the retransmission if an error occurs. The above module is implemented to build health data management systems for user to retrieve their own health data. At last, implement SUS test to assess the system usability. In this study, we have implemented the health management system. The simple operation interface allows users to carry out equipment pairing, received data and parameter real-time presentation. Simulation of random error for data transmission to test error detection rate, and the test results get nearly 100% error detection rate. Cyclic redundancy check code with high efficiency characteristics can reduce the burden of error detection. The SUS test of system usability shows that the user is generally satisfied with the concise interface and operation flow of the system, but the satisfaction of the system in “relation to the system” is relatively low. Therefore, it is possible to further understand whether there is a function that the user expects or should have but not integrated into the system. For the future of development, we can conduct assessment in the data transmission error prevention module of the benefits of integrating the error correction code and the appropriateness of network situation. For the application of data can be combined with advanced data exploration methods to provide more flexibility and personalization analysis services.