透過您的圖書館登入
IP:3.140.185.147

摘要


根據歷年資料的統計,許多居家照護的死亡原因之中以意外傷害居多,而其中大部分原因往往是家人或照護人的一時疏忽而釀成了大錯,因而傷害了一個脆弱的生命,且隨著全球高齡化趨勢的發展及少子化現象,可以預見高齡化趨勢勢必衍生出許多健康照護方面的問題及慢性病患人數快速增加和藥物的需求。而目前市面上的的照護系統皆需透過大量人力操作及昂貴的儀器設備,加上人口少子化的影響,可能造成未來一名照護員必須同時間照護數十名老人的情況。因此本論文提出一套結合多項生理訊號感測裝置及視覺辨識技術之健康監控系統,裡面結合RFID技術、血氧脈搏檢測器、人臉表情偵測及年齡辨識等儀器和技術,利用嵌入式系統將感測到的所有資訊作決策判斷,若有危險的狀況產生,系統便會發出警訊通知相關人員,如此可以減少照護人員的辛勞並提供更完善的照護給被照護者。從實驗結果可以了解我們所提出來的系統是一個確實可以被實踐和被使用的系統,且在系統中我們所提出的人臉表情辨識和人的年齡偵測技術比起其他方法有更好的辨識與偵測的結果。

並列摘要


Based on the statistics of previous years, a major reason for death occurring in patients being cared for at home and residential assisted living facilities is accidental, often resulting from negligence of the relatives or caregivers and resulting in harm to already fragile human beings. As populations continue to age and the birth rate shrinks, the number of chronic patients is expected grow, a rapidly increase in need for drugs for the chronically ill will occur, as well as the increase in accidents related to future lack of the human resources to monitor and care for these patients and resulting in an increase in accidental deaths. These future problems will be worsened if we do not update the current health monitoring system, which is operated manually, which makes use of expensive equipment, and which is performed quite slowly. Given the increases in the chronically ill ages, it is likely in the future that dozens of elderly people will just have one caregiver among them.. Hence, this paper proposes a multifunctional health monitoring system to provide better and safer monitoring than traditional human care. The system includes such hardware as RFID devices to monitor hand temperature, ambient temperature and humidity, a blood pressure and pulse detector to monitor blood pressure and pulse, and a webcam to monitor human expression and age. All information collected by such sensors can be integrated and processed with an embedded system. Using a rule-based decision method, such a system should be able to evaluate a patient's state of health. Should the system detect an unstable status, it can send warning messages to related staff caretakers who could then take the necessary actions to care for the patient in distress in a timely manner. This system would save patient-to-patient monitoring time and allow caretakers to provide immediate care to those in more urgent need of it. Our tests of this proposed system suggest that that it is possible to set up a complete health monitoring system by combining this embedded system and system-alerted health-care service.

延伸閱讀