銀髮族智慧照護與商業模式 －跌倒狀態評估系統之應用

Smart-Care in Senior Citizens and Business Model: Applying Fall State Evaluation System

莊秀滿(Hsiu-Mang Chuang)

指導教授 ： 李正文

中原大學/商學院/企業管理學系/博士(2021年)

https://doi.org/10.6840/cycu202201010

摘要

銀髮族最常見的就是遇到跌倒問題，往往造成非常嚴重的傷害。根據國民健康署在2017年「國民健康訪問調查」3,280位65歲以上老人中，自述過去一年曾跌倒者就有495人(15.5%)，也就是說每6個老人，就有1位在一年內跌倒的經驗，在107年死因統計中，跌倒高居65歲以上事故傷害死亡原因之第二位(每十萬人25.7人)。嚴重跌倒可能會造成患者長期臥床，甚至威脅高齡者生命危險。
健康醫療網研究指出，失智症族群及高齡族群所需之服務比其他家人更為殷切，每年約有三分之一的65歲以上居家老人發生跌倒的意外傷害事故。由於老人隨著年齡增加，居家或在安養機構因跌倒傷害機率亦隨之增加且嚴重(衛福部2015年事故傷害公布資料)。依據臺灣老年醫學會會訊"老年人跌倒的流行病學和危險因子的評估和預防"指出，老年人跌倒常見的危險因子可歸納為內在因素和外在因素，外在因素包括環境等因素，內在因素包括因年齡增加所產生的生理功能的退化、急慢性疾病和藥物，其中急性疾病如心肌梗塞、心律不整、心臟衰竭、腦中風、神智混亂、癲癇發作、貧血、電解質不平衡等病症是造成跌倒的危險因子。因此，若能從平日或跌倒發生時獲得危險因子之生理資訊，可以做為醫護人員的重要參考。
老人跌倒危險因子很多，故本研究以重要項目優先進行分析以整合跌倒狀態評估服務，讓照護人員透過平台方便照護及監控異常，有效降低配戴者的風險。
根據警政署2018年公安統計資料指出，近3年來平均每年都有逾3400 老人失蹤，每天近10 位老人失蹤，台灣失智症協會指出，走失事件頻傳是失智人口增加過程可預期的現象，且未來還會增加速度還會更快。根據台大醫院神經部研究，失智老人在走失48~72 小時仍未被尋獲的話，死亡率高達4 成。
本研究的主要目的是構建一個平台和相關算法，從而提供跌倒狀態評估，並使用集成了加速度計和陀螺儀的智能盒子佩戴在腰間，以達到獲得完整、穩定的測量信息的目的。通過手機應用程序，親屬和監護人可以通過APP查看分析結果，實現快速檢測，高效幫助患者。通過智能盒子的使用，結合定位和跌倒狀態分析，未來有望提供與居家服務和長照機構合作的一體化接口，適用於居家和機構護理單位。因此，這是本文研究的主要動機。
其主要功能包括：
1） 使用集成了加速度計和陀螺儀的智能盒子，佩戴在腰間，達到獲得完整穩定
測量信息的目的。
2） 使用跌落狀態評估算法來提高準確性。
3） 在後台使用機器學習算法分析佩戴者的使用信息，提供適應性閾值並更新前端硬件以適應用戶群體。
4） 按照動作安全載體的服務理念設計，整合目前的定位和跌倒狀態分析服務，結合更多可穿戴式看護設備。
在本研究中，針對老年人面臨的問題，結合未來行動安全護理服務的概念和商業應用，提供適當的護理服務。

關鍵字

跌倒 ； 老年人 ； 行動安護 ； 定位

並列摘要

The seniors are prone to fall down and which may cause seriously hazard to them. In accordance with the National Health Interview Survey conducted by the National Health Administration in the year of 2017, 495 (15.5%) of 3280 seniors at the age of 65 years old or above told that they had the experience of falling down in the past one year (Health Promotion Administration(MOHW),National Health Interview Survey in 2016), namely 1 of 6 seniors had the experience of falling down in one year; in the death causes statistics made in the year of 2018, the fall-down ranked top 2 among the death causes and accidental hazard happened to the seniors at the age of 65 years old (25.7 of 100,000 persons). The serious fall may result in the patient to suffer with the long-term bed or even threaten the life of seniors.
As what it was pointed out by the researches as posted on the Health and Medicine Network, the people with dementia and the seniors are more urgent for the related services than other families, approximately more than 1/3 in-home elderly at the age of 65 years old are prone to fall down and may suffer with the accidental injury every year (healthnews , 2017). As the age increases, the probability that the seniors at home or in the pension agency may fall down and suffer with hazard therefrom increases too and even becomes more serious (data of accident injury published by Health and Welfare Ministry 2015). According to the news of Taiwan Association of Gerontology and Geriatrics, namely “Evaluation and Prevention on Fall-down Epidemiology and Hazardous Factors of Seniors”, the common hazardous factors for fall of senior can be classified into the internal factors and the external factors, where the external factors include environmental factors, the internal factors include the degeneration in the physiological function as the age increases, acute or chronic disease and their drugs, where the acute diseases such as myocardial infraction, cardiac arrhythmia, cardiac failure, cerebral apoplexy, mental disorder, epileptic seizure, anemia, electrolyte imbalance and other diseases are the dangerous factors causing fall. Therefore, if the physiological information of risk factors can be acquired on every ordinary day or when fall happens, it can be used as the important reference for the medical workers.
There are many risk factors for fall of seniors. In this work, we gave priority to the important items for making analysis so as to integrate the fall-status evaluation services and let medical workers start easy care and monitor abnormality via the platform, and finally reduce efficiently the risks of wearers.
According to the statistical data on Public Security of Police Administration 107, there were almost 3,400 old persons missing every year in recent three years, namely almost 10 old persons were missing every day (healthnews , 2014). As what it was pointed out by Taiwan Alzheimer Disease Association, the frequently happened missing event is an expectable phenomenon in the process when the populations with Alzheimer Disease increase, and the increase will expedite in future. In accordance with the research of the Nervous Department of NTU Hospital, the senior with the Alzheimer Disease would be died in 48~72 hours after missing and their death rate would be 40%.
The main purpose of this research is to construct a platform and related algorithms and thereby provide a fall-state assessment, and use the smart box integrated with accelerometer and gyroscope and wear it around the waist to achieve the purpose of achieving complete and stable measuring information. With the application programs of mobile phone, the relatives and guardians can check analysis results via APP and realize quick detection and efficient help to patients. With the use of smart box, and by integrating with positioning and fall-state analysis, it is expected to provide the integrated interfaces to cooperate with home service and long-term care agencies in future, which is suitable for home and institutional care units. Therefore it is the main motivation of the research in this work.
Its main functions include:
1) Use the smart box integrated with accelerometer and gyroscope and wear it around
the waist to achieve the purpose of achieving complete and stable measuring
information.
2) Use the fall-state evaluation algorithm to improve accuracy.
3) Use the machine learning algorithm in the background to analyze the use information of wearer, provide adaptability threshold and renew front-end hardware to suit for the user populations.
4) Design according to the service concept of action safety carrier to integrate the positioning and fall-state analysis service at present and combine more wearable care-taking equipment .
In this study, appropriate care services in accordance with the problems confronted by the seniors and by combining the concept of action safety care service in future and commercial applications.

並列關鍵字

fall ； senior (old person) ； action safety care ； positioning.

參考文獻

Bagala, F., Becker, C., Cappello, A., Chiari, L., Aminian, K., Hausdorff, J. M., ... & Klenk, J. (2012). Evaluation of accelerometer-based fall detection algorithms on real-world falls. PloS one, 7(5), e37062.

Bosch Sensortec Community. BMI160 IMU spec. https://community.bosch-sensortec.com/t5/Knowledge-base/BMI160-Series-IMU -Design-Guide/ta-p/7376.

Boren, Y. (2012). Predictors of Fall among The Elderly in Taiwan. Degree Thesis of Institute of Medicine, Sun Yat-Sen Medical University, (1-99).

Broadley, R. W., Klenk, J., Thies, S. B., Kenney, L. P., & Granat, M. H. (2018). Methods for the real-world evaluation of fall detection technology: A scoping review. Sensors, 18(7), 2060.

Casilari, E., Santoyo-Ramón, J. A., & Cano-García, J. M. (2017). Umafall: A multisensor dataset for the research on automatic fall detection. Procedia Computer Science, 110, 32-39.