CSE 5810 Sensor Networks to Monitor Elderly Yusuf

CSE 5810 Sensor Networks to Monitor Elderly Yusuf Albayram Computer Science & Engineering University of Connecticut, Storrs yusuf. albayram@uconn. edu Albayram-1

Introduction m CSE 5810 m m The proportion of elderly in the world is demonstrating a remarkable increase every year. q By the year 2050, 1 in 5 person in the world will be age 60 or older, 1. 6 million people in the aging population live in facilities Typical residents need assistance with 2 activities of daily living Albayram-2

Problems m CSE 5810 With the increase of elderly people population: q Rising Health Care Costs Ø More investment is needed for elderly care q Many elderly people choose to stay at home Ø e. g. , Due to privacy/dignity issues. q A majority of older adults are challenged by chronic and acute illnesses and/or injuries. Ø 80% of older Americans have one or more chronic diseases. q The growing insufficiency of traditional family care Ø i. e. , decreased care by relatives q Decrease in the working population will cause a shortage of skilled caregivers. Albayram-3

State of the art applications m CSE 5810 m Advances in sensor technology, object localization, wireless communications technologies can q enable elderly people to regain their capability of independent living q make possible unobtrusive supervision of basic needs of frail elderly and thereby replicate services of on-site health care providers Assisted Living Technologies are expected to contribute significantly q improving the quality of life of elders q reducing costs by avoiding premature institutionalization Albayram-4

What services can assisted living systems offer? m CSE 5810 m m m Alarms/notifications and triggers Queries Reminders Detect anomalies and deviations Recognize specific behaviors and assist with task completion Keep the person active and connected to the social environment Albayram-5

Overview m Introduction & Motivation m Sensor Networks to Monitor Elderly q (1) Activities of Daily Living Monitoring, q (2) Location Tracking, q (3) Medication Intake Monitoring, q (4) Medical Status Monitoring, q (5) Fall and Movement Detection m Challenges CSE 5810 Albayram-6

(1. 1) Activities of Daily Living Monitoring m CSE 5810 m Monitoring the patient’s activities of daily living (ADLs) is essential to q Detects anomalies and prompts them, q Assist the independent living of older adults q The diagnosis of diseases and health problems Several projects have investigated the use of pervasive sensors to provide a ‘smart’ environment for the observation of (ADL) q The use of heterogeneous sensors, including Ø Wearable sensors (Body Sensor Network (BSN)) – Designed to collect biomedical, physiological and activity data Ø Ambient sensors (Ambient Sensor Network (ASN)) – Designed to collect data around the region where the ADL takes place. Albayram-7

(1. 2) Activities of Daily Living Monitoring m CSE 5810 Variety of multi-modal and unobtrusive wireless sensors seamlessly integrated into ambientintelligence compliant objects (AICOs) to achieve activity recognition [17] Overview of assisted living populated with a variety of wireless multimodal sensors to collect data for various ADLs Albayram-8

(2) Location Tracking m CSE 5810 m m m 25% of people over 60+ suffer from Alzheimer’s and Dementia Seniors with Dementia or Alzheimer’s can easily become confused or lost. Monitoring location of a person suffering dementia or Alzheimer’s can help q Detect signs of disorientation or wandering. q The health professional to reach a diagnosis of a type of dementia. Several methods for location tracking have been proposed: q (1) GPSs based outdoor location tracking q (2) RFID-based indoor location tracking q IR, ultrasound Albayram-9

(2. 1) Location Tracking m CSE 5810 (1) GPSs based outdoor location tracking q GPS-enabled devices include an SOS button and once pressed , connect with their family member or caregiver. GPS Tracker Bracelets Wearable AGPS terminal Smart Phone with GPS Albayram-10

(2. 2) Location Tracking m CSE 5810 m m (2) RFID-based indoor location tracking GPS does not work in indoor Real-time monitoring of elderly people’s whereabouts q The movement of the elderly person wearing an RFID tag is sensed by the RFID readers installed in the building The RFID-based location sensing system in smart home environments Albayram-11

(2. 3) Location Tracking m CSE 5810 Critique for location tracking systems q Privacy is one of the major issue q Too battery-hungry and battery drain quickly (e. g. , smart phones) q Devices must be lightweight, small, and comfortable to wear and use q Elders often have no idea using computers, smartphones and other technological tools Ø their interaction with them must be simple Ø And limited to a minimum Albayram-12

(3) Medication Intake Monitoring m CSE 5810 m Taking medications is one of the most important activities in an elder’s daily life q Elders taking on average of about 5. 7 prescription medicines and 4 nonprescription drugs each day [15] Medication intake monitoring is essential q m Medication noncompliance is common in elderly and chronically ill especially when cognitive disabilities are encountered [13]. The existing methods/systems often utilize following sensor technologies for medication intake monitoring : q RFID q Computer vision Albayram-13

(3. 1) Medication Intake Monitoring m CSE 5810 Integrating both sensor network and RFID technologies q HF RFID tags to identify when and which bottle is removed or replaced by the patient q The weight scale monitors the amount medicine on the scale q The patient wearing an Ultra High Frequency (UHF) RFID tag is determined in the vicinity and alert the patient to take the necessary medicines. Medicine Monitor System Prototype Albayram-14

(3. 2) Medication Intake Monitoring m CSE 5810 Incorporating RFID and video analysis [10] q RFID tags applied on medicine bottles located in a medicine cabinet and RFID readers detect if any of these bottles are taken away q A video camera monitoring the activity of taking medicine by integrating face and mouth detection Monitoring the activity of taking medications using computer vision-based method RFID system includes antenna and RFID reader Albayram-15

(4) Medical Status Monitoring m CSE 5810 Health monitoring devices are primary responsible for q Collecting physiological data from the patient Ø (e. g. , ECG, heart rate, blood pressure) Transmitting them securely to a remote site for further evaluation At the health provider’s end, q the medical personnel and supervising physicians can have instant access to q m Ø real-time physiological measurements Ø the medical history of several monitored patients Albayram-16

(4. 1) Medical Status Monitoring CSE 5810 The health monitoring network structure [16] Albayram-17

(5) Fall and Movement Detection m CSE 5810 m m Fall Events very common situation in elderly people q 30% of the older persons fall at least once a year q Fall responsible of 70% of accidental death in persons aged 75+ There are primarily 3 types of fall detection methods for elderly q (1) Wearable device based methods q (2) Vision based methods q (3) Ambient based methods Once the fall event was detected, an alert email is immediately sent to the caregiver Albayram-18

(5. 1) Fall and Movement Detection m CSE 5810 (1) Wearable device based methods q Using accelerometers and gyroscopes to analyze changes in a body’s position to detect falls. the sensor nodes are attached on the chest (Node A) and thigh (Node B) A tri-axial accelerometer for monitoring acceleration and a tri-axial gyroscope for monitoring angular velocity [14] Albayram-19

(5. 2) Fall and Movement Detection m CSE 5810 (2) Vision based methods q Detect Fall from a video sequence by: Ø Applying background subtraction to extract the foreground human body and post processing to improve the result [2, 3] Albayram-20

(5. 3) Fall and Movement Detection m (3) Ambient based methods q Rely on pressure sensors, acoustic sensors or even passive infrared motion sensors, which are usually implemented around caretakers’ houses m Once the fall event was detected, an alert call/email was immediately sent. CSE 5810 Albayram-21

(5. 4) Fall and Movement Detection m CSE 5810 Critique for automatic fall detection, q (+) Video based methods are usually more accurate q (-) Video based methods raise privacy concerns q (+) Acoustics based methods are very susceptible to ambient noise q (-) Video-based and acoustic-based methods are costly due to pre-installation q (-) Wearable based methods operate as long as the person wears the sensors q (+) With the improvements in smart phone tech (built-in sensors e. g. , accelerometer, gyroscope), Smart phones are ideal for developing an app that can automatically detect falls and provide a warning mechanism. Albayram-22

CSE 5810 Challenges of Sensor Networks solutions for monitoring elderly m Hardware level challenges q Unobtrusiveness q Sensitivity and calibration q Energy q Data acquisition efficiency m Security m Privacy m User-friendliness m Ease of deployment and scalability m Mobility Albayram-23

References CSE 5810 [1] Wang, J. , et al. "An enhanced fall detection system for elderly person monitoring using consumer home networks. " Consumer Electronics, IEEE Transactions on 60. 1 (2014): 23 -29. [2] Yu, Miao, et al. "A posture recognition-based fall detection system for monitoring an elderly person in a smart home environment. " Information Technology in Biomedicine, IEEE Transactions on 16. 6 (2012): 1274 -1286. [3] Foroughi, Homa, Baharak Shakeri Aski, and Hamidreza Pourreza. "Intelligent video surveillance for monitoring fall detection of elderly in home environments. " Computer and Information Technology, 2008. ICCIT 2008. 11 th International Conference on. IEEE, 2008. [4] Yavuz, Gokhan, et al. "A smartphone based fall detector with online location support. " International Workshop on Sensing for App Phones; Zurich, Switzerland. 2010. [5] Popescu, Mihail, et al. "An acoustic fall detector system that uses sound height information to reduce the false alarm rate. " Engineering in Medicine and Biology Society, 2008. EMBS 2008. 30 th Annual International Conference of the IEEE, 2008. [6] Huang, Kung-Ta, et al. "An intelligent RFID system for improving elderly daily life independent in indoor environment. " Smart homes and health telematics. Springer Berlin Heidelberg, 2008. 1 -8. [7] Ferreira, João. "Behavioral Analytics for Medical Decision Support: Supporting dementia diagnosis through outlier detection. " (2012). [8] Wong, AK-S. , et al. "An AGPS-based elderly tracking system. " Ubiquitous and Future Networks, 2009. ICUFN 2009. First International Conference on. IEEE, 2009. [9] Kim, Soo-Cheol, Young-Sik Jeong, and Sang-Oh Park. "RFID-based indoor location tracking to ensure the safety of the elderly in smart home environments. " Personal and ubiquitous computing 17. 8 (2013): 1699 -1707. [10] Hasanuzzaman, Faiz M. , et al. "Monitoring activity of taking medicine by incorporating RFID and video analysis. " Network Modeling Analysis in Health Informatics and Bioinformatics 2. 2 (2013): 61 -70. Albayram-24

References-2 CSE 5810 [11] Pang, Zhibo, Qiang Chen, and Lirong Zheng. "A pervasive and preventive healthcare solution for medication noncompliance and daily monitoring. " Applied Sciences in Biomedical and Communication Technologies, 2009. ISABEL 2009. 2 nd International Symposium on. IEEE, 2009. [12] Ho, Loc, et al. "A prototype on RFID and sensor networks for elder healthcare: progress report. " Proceedings of the 2005 ACM SIGCOMM workshop on Experimental approaches to wireless network design and analysis. ACM, 2005. [13] Alemdar, Hande, and Cem Ersoy. "Wireless sensor networks for healthcare: A survey. " Computer Networks 54. 15 (2010): 2688 -2710. [14] Li, Qiang, et al. "Accurate, fast fall detection using gyroscopes and accelerometer-derived posture information. " Wearable and Implantable Body Sensor Networks, 2009. BSN 2009. Sixth International Workshop on. IEEE, 2009. [15 ] Johnston C (2001) Falls in the Elderly, UCSF Division of Geriatrics Primary Care Lecture Series. http: //s 3. amazonaws. com/engrademyfiles/4063195431780411/sf_falls. ppt [16] Pantelopoulos, Alexandros, and Nikolaos G. Bourbakis. "Prognosis—a wearable healthmonitoring system for people at risk: Methodology and modeling. " Information Technology in Biomedicine, IEEE Transactions on 14. 3 (2010): 613 -621. [17] Lu, Ching-Hu, and Li-Chen Fu. "Robust location-aware activity recognition using wireless sensor network in an attentive home. " Automation Science and Engineering, IEEE Transactions on 6. 4 (2009): 598 -609. Albayram-25