進階搜尋


 
系統識別號 U0026-2704202016283200
論文名稱(中文) 開發一套智慧病房監測系統
論文名稱(英文) Development of a Smart Ward Monitoring System
校院名稱 成功大學
系所名稱(中) 生物醫學工程學系
系所名稱(英) Department of BioMedical Engineering
學年度 108
學期 2
出版年 109
研究生(中文) 江柏楷
研究生(英文) Bo-Kai Chiang
學號 P86071228
學位類別 碩士
語文別 英文
論文頁數 50頁
口試委員 指導教授-陳天送
口試委員-陳琮琳
口試委員-黃至誠
口試委員-閻漢琳
口試委員-陳啟杰
中文關鍵字 智慧病房監測  Wi-Fi通訊技術  輕量級網路協定  IoT  MQTT Broker  發布/訂閱 
英文關鍵字 Smart ward monitoring  Wi-Fi communication technology  lightweight network protocol  IoT  MQTT Broker  Publisher / Subscriber 
學科別分類
中文摘要 現今社會已邁入人口高齡化,醫院的醫療照護負擔將隨之增加,一旦人力匱乏,病患發生意外的比例將大幅上升。根據衛生福利部統計,2018年台灣已邁入高齡化社會,至2026將成為超高齡化社會,亦即每10人就有4位是65歲以上的老年人。據衛福部統計,醫院的意外事故多以跌倒居多,其中跌倒地點又以病房為主,雖然透過加裝攝影機可以監測病房內的狀況,但使用攝影機仍有侵犯隱私的疑慮。因此,開發一套智慧病房監測系統,減少醫護人力負擔、全天候監控且沒有隱私侵犯的疑慮,將會是提升醫院照護效率與安全性的重要項目。
跌倒事故一旦發生,被護士發現時可能已經過一段時間,如錯過危及處理的黃金期,恐會導致更嚴重的後果。目前常見的警報裝置,如設置在廁所的緊急按鈕,如果患者跌倒在廁所中或是陷入昏迷狀態,就無法按下緊急按鈕來求救。在此,這個領域確實沒有相關的系統來確保患者在病房內的安全,尤其是在醫療照護人力缺乏的時候。不過,智慧病房監測系統可以使護理人員透過護理站的電腦遠端掌握病房內患者的行為,並詳細記錄活動空間及其時間,此系統的功能有助於改善目前的情況。
該研究提出了一套智慧病房監測系統,可遠端監控且沒有任何侵犯隱私的問題,該系統將病房分成多的空間,例如分成病床、病房內、廁所與房門,透過監測裝置偵測目前使用者所處的空間,來掌握使用者的狀況。該系統使用一組兩對超音波感測器作為進入或離開廁所或房間的判斷裝置,以及一個壓力感測器作為偵測使用者上床或下床的判斷裝置,同時系統將各點位的監測裝置透過Wi-Fi通訊技術連結上醫院的Wi-Fi,並將各裝置的偵測資料上傳至使用MQTT網路協定的伺服器,該伺服器為病房裝置與護理站電腦的資料傳輸媒介。最後,護理站的醫療人員可透過站內電腦的監控介面來掌握病患在病房內的活動情形與進出個空間的詳細時間。
測試結果顯示,超音波感測裝置為了減少偵測時重疊區域所造成的誤判,兩顆超音波探頭必須分別向外旋轉大約10度來改善此問題,以及,測試的區域需要距離設備20 cm以避開死角。為了使壓力感測裝置的偵測更為準確,重量低於40 kg的使用者在上下床時,須從安裝監測裝置側下床,以確保壓力感測器的壓力變化量夠大,足以判斷使用者的上下床。使用MQTT協定的主題做資料的發布與訂閱時,原本採用多個主題的方式分別傳輸,但因為多個客戶端及多個主題同時於伺服器做發布與訂閱,導致伺服器傳輸速率產生延遲甚至錯誤,因此在伺服器傳輸速率有限的情況下,須改以三組裝置對應一組接收器,伺服器也僅使用一個主題來提供發布與訂閱。最後,智慧病房監測系統的技術不僅可以使用在醫院護理站與病房,該設備還可應用於行動不便的人或獨居老人的家庭安全。
英文摘要 The society has entered an aging population, and the burden of hospital nursing staff will increase greatly at the same time. Once there is a lack of nursing staff resource, the rate of patients with incidents will increase significantly. According to statistics from the Ministry of Health and Welfare, in 2018, Taiwan has a trend in entering an aging society, and then it will become a super-aged society, with 4 out of 10 people being elderly over 65s in 2026. According to the statistics of the Ministry of Health and Welfare, most incidents in hospitals are caused by falling over, and the places are mainly ward. Although the camera can be installed to monitor the conditions in the ward, the use of cameras still has privacy concerns. Therefore, the development of a smart ward monitoring system to reduce the burden of medical staff, long-term monitoring, and no privacy violations will be an important project to improve the efficiency and safety of hospital care.
Once the incident of falling occur in hospital, it may have been a while before being discovered by the nurse, which may lead to a serious consequence. The common alarm devices such as the emergency button installed in the bathroom, but if the patient falls over or falls into a coma, he can’t press the emergency button to call for help. In this circumstance, there is no such a system can ensure the safety of patients, especially when the nursing staff is lacking. However, the smart ward monitoring system allows nursing staff to monitor the locations of patients through the computer of the nursing station then record the time in detail.
The study proposes a smart ward monitoring system without any privacy violations. The ward is divided into multiple areas, such as the bed, the inner space, a bathroom and the doorway, and get the user's status by detecting the area. The system uses two pairs of ultrasonic sensors to detect whether someone enters or leaves the bathroom or room, and one pressure sensor to detect whether someone is on the bed or out of bed. Then, the monitoring devices use Wi-Fi communication technology to connect to the local area network of hospital, and upload the data received from devices to the server that using the MQTT protocol. The server is a broker of the data transmission broker of monitoring device and the computer of the nursing station. After that, the computer interface monitors the patient's location and records the time point in the ward.
The test results show that in order to reduce the misjudgment from overlapping area in detection, the two sets of sensors must be rotated outward by approximately 10 degrees. In addition, the test area needs to be 20 cm away from the device to avoid blind spots. To make the detection of the pressure sensor more accurate, users under 40 kg need to get out of the bed from the side where the sensor is installed to ensure that the pressure change is large enough to judge whether he is on be or not. At the beginning, the system uses the MQTT broker’s topic to publish and subscribe, it used multiple topics to transmit data in separate, but since there are multiple clients publish and subscribe to multiple topics in one single server, the transmission rate causes delays or even errors. Since the transmission rate of the server is limited, the three devices must match only one receiver, and the server also uses only one topic to provide publish and subscribe. Finally, the technology of the smart ward monitoring system can be used not only in hospital nursing stations and wards, but also in the home safety of people with reduced mobility or elderly people who living alone.
論文目次 摘要 I
Abstract III
致謝 V
List of Figures VIII
List of Tables X
Chapter 1 Introduction 1
1.1 World Population Ageing 1
1.2 The Impact of Taiwan Population Aging 3
1.3 Hospital Incident and Fall Over 5
1.4 Literature Review 8
1.4.1 Continuous Monitoring 8
1.4.2 Cloud Healthcare Data Storage 9
1.5 Motivation and Aims 10
Chapter 2 Materials and Methods 11
2.1 System Architecture 11
2.2 Ward Area Detection-Get In and Out of Bed 13
2.2.1 Force Sensing Resistor 14
2.2.2 Force change Algorithm 16
2.3 Ward Area Detection-Door Passing 18
2.3.1 Ultrasonic Sensor 19
2.3.2 Dual Probe Detection Algorithm 20
2.4 Nursing Station-Detection Data Receiver 23
2.4.1 Lightweight Network Protocol-MQTT 24
2.4.2 Data Transmission Process 26
2.5 Nursing Station-Monitoring Interface 28
2.6 Development Board 31
Chapter 3 Results and Discussion 35
3.1 Algorithm Testing 35
3.1.1 Force Test of Body Position on Bed 35
3.1.2 Calibration of Ultrasonic Probe 36
3.2 Accuracy 39
3.3 Connection Quality of MQTT Broker 42
3.3.1 MQTTLens Plugin 42
3.3.2 Data Transmission Simplification 44
3.4 Interface Optimization 45
3.4.1 Input Unit Simplification 45
3.4.2 Output Display Refinement 46
Chapter 4 Conclusion 47
References 48

參考文獻 [1] World Health Organization, Ageing and life course infographics, Populations are getting older, 2020.
[2] Taiwan’s National Development Council, Population Estimation Inquiry System, Aging indicators graphic, 2018.
[3] Prof Linda H Aiken, Douglas M Sloane, Luk Bruyneel, Koen Van den Heede, Prof Peter Griffiths, Prof Reinhard Busse, Nurse staffing and education and hospital mortality in nine European countries: a retrospective observational study, The Lancet, February 26, 2014.
[4] Ministry of Health and Welfare, hospital incidents, 2017.
[5] Ministry of Health and Welfare, the place of falling over, 2017.
[6] Ministry of Health and Welfare, the situations of falling over, 2017.
[7] Ministry of Health and Welfare, the elders living companion and alone, 2017.
[8] Hao, J., Jayachandran, M., Kng, P.L., Foo, S.F., Aung, P.W., & Cai, Z, FBG-based smart bed system for healthcare applications. Frontiers of Optoelectronics in China, 3, 78-83, 2010.
[9] Thomas, A.M., Moore, P., Shah, H., Evans, C., Sharma, M., Xhafa, F., Mount, S., Pham, H.V., Wilcox, A.J., Patel, A., Chapman, C., & Chima, P, Smart care spaces: needs for intelligent at-home care. IJSSC, 3, 35-44, 2013.
[10] Intelligent Inpatient Ward Solution, System, gvssmart.com, 2018.
[11] Tekscan, FlexiForce, model A201 Sensor.
[12] FlexiForce, product catalog, Tekscan.
[13] FlexiForce, FLX-FlexiForce-Sensors-Manual, Tekscan.
[14] Roderick Burnett, Understanding How Ultrasonic Sensors Work, MaxBotix, 24-03-2020.
[15] MB1030 LV-MaxSonar-EZ3, MaxBotix ultrasonic sensor product catalog.
[16] R. DePaula, L. Flax, J. Cole, J. Bucaro, Single-mode fiber ultrasonic sensor, IEEE Journal of Quantum Electronics, 1982.
[17] Archer, 運作方式, [ Protocol ]認識MQTT, 2019.
[18] cubie, 比較HTTP和MQTT通訊協定, MQTT教學(一):認識MQTT, 2017
[19] ARDUINO NANO 33 IOT, store.arduino.cc, 2020.
[20] ARDUINO NANO 33 IOT, DOCUMENTATION, store.arduino.cc, 2020.
[21] Marcia Kilchenman O'Malley, Calibration of the FSR voltage with force measured, ResearchGate, Oct 2013.
[22] Kai-Tai Song, Chih-Hao Chen, Cheng-Hsien Chiu Huang, Design and experimental study of an ultrasonic sensor system for lateral collision avoidance at low speeds, IEEE Intelligent Vehicles Symposium, 2004.
[23] S. Saitoh, M. Izumi, Y. Mine, A dual frequency ultrasonic probe for medical applications, IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, Volume: 42, Issue: 2, March 1995.
[24] Test connection using MQTTLens ,litmusautomation.com,2013-2018.
[25] Shinho Lee, Hyeonwoo, Kim Dong-kweon, Hong Hongtaek Ju, Correlation analysis of MQTT loss and delay according to QoS level, The International Conference on Information Networking, 2013.
[26] Y. Imai, M. Ooga, D. Yamane, O. Sadayuki, Y. Iwamoto, S. Masuda, Mobile phone-enhanced user interface of remote monitoring system, IEEE International Conference on Mobile Business, 2005.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2020-08-18起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2020-08-18起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw