用非接觸式感應技術做智慧型病房內意外事件之偵測系統

跌倒是影響老年人健康及生命的重要危險因數，根據調查顯示，跌倒占居家活動中意外的34.5%。跌倒除了危害生命安全，亦會增加社會醫療費用與家庭照顧成本的負擔。隨著台灣進入高齡化社會，老年人口日漸增多，跌倒的防範與即時偵測已成為現代預防醫學及家庭健康照護之重要課題。本論文利用非接觸式感應技術，以病人為中心，做病房內各種活動之跌倒意外事件偵測，目的是為提昇醫療服務品質，減低意外的危險。傳統的居家照護方法需仰賴人力全天後的照顧，大多數的家屬也無法24 小時伴隨在病患身旁。無法照護的時間，得仰賴醫護人員的照顧，但隨著工資的逐年上漲，人事支出是一筆很大的負擔。而一般的居家照護系統，則是利用在病患衣物內裝置微感應設備進行意外事件的偵測。其必須仰賴穿戴在身上的感應設備，相對的也造成病患行動的不便。本論文提出一套居家及病房的跌倒意外事件偵測系統，病患可以不用穿戴厚重的感應設備，便能達到跌倒防護的目的。本系統以網路攝影機進行病房內跌倒偵測，藉由分析病房內病患的活動軌跡，進行跌倒行為的偵測。在不便攝影機拍攝的地方，像是洗手間，則利用無線射頻技術進行停留時間的異常偵測。系統如果偵測到跌倒行為或停留過久之異常狀況時，便即時通報醫護人員避免延誤就醫。透過本系統可減輕醫護人員及醫療成本的負擔。吟

關鍵字

電腦視覺、意外事件偵測、跌倒。

並列摘要

Fall injuries strongly affect the health of senior citizen, and place lives in serious danger. Fall injuries account for 34.5% of accidents at home, according to the results of investigation. As well as endangering safety, falls increase the medical expenses borne by society and families. As Taiwanese society ages, with the older population continuously increasing, detecting and preventing falls becomes increasingly important for protecting family health. This thesis presents a non-contact patient-centered sensing technology to detect and examine accidental falling events. The aim of the proposed system is to improve medical treatment and reduce unexpected danger. Protection in the home requires 24-hour care. This care must be provided by medical personnel. Expenditure from wages of medical personnel is an increasing burden. Therefore, the proposed general home care system places micro sensor devices into a patient's body and limbs to detect and examine accidents. The sensor devices need to be placed on the body causing patient is inconvenient. This thesis develops an accidental event detection system for home and ward. Patients can achieve protection from falling without wearing heavy sensor devices. Falls are detected by network cameras in the hospital ward to analyzing the patient's activity. If cameras cannot be used, such as bathroom, then RFID (Radio Frequency Identification) technology can be used instead. If the system detects the falling or other unusual behavior, then the medical personnel are notified immediately to seek medical advice. This system can be applied to reduce the cost and burden to medical care personnel.t

並列關鍵字

computer vision, accidental event detection, fall

參考文獻

persons living in the community”, New England Journal of Medicine, 319,

1701-1707, 1988.

[4] Haritaoglu, D. Harwood and L. S. Davis, “W4：who? when? where? what? a

Computer Vision, Vol. 579, 1985, pp. 72–78.

[6] J. Barron, D. Fleet and S. Beauchemin, “Performance of optical flow

國際替代計量

用非接觸式感應技術做智慧型病房內意外事件之偵測系統

未授權

主題瀏覽