The Use of Smart Watches to Monitor Heart Rates in Elderly People PDF

Preview

Summary

hhhhhh...

Description

The Use of Smart Watches to Monitor Heart Rates in Elderly People: A Complementary Approach Majid H. Alsulami1&2 College of Community, Shaqra 1 Shaqra University Saudi Arabia +44 (0) 7738688838 [email protected]

Anthony S. Atkins 2 Faculty of Computing, Engineering and Sciences 2 Staffordshire University Beaconside, Stafford, UK ST18 0AD +44 (0) 1785 353453 [email protected]

Abstract— Heart failure is a significant concern for elderly people. Serious problems can result if elderly people do not take appropriate action to prevent medical issues. As a solution, smart watches can be used to monitor heart rate. These devices can also allow families, relatives, friends and health-care providers to visualise the heart rate data via a smartphone application or online website. Smart watches have limited functionality, and experiments were conducted to test the functionality they provide to elderly persons, including heart rate monitoring. The results show that there is currently no proactive action that can be taken when an abnormal heart rate occurs. This paper proposes a complementary approach using a knowledge-based system with rule-based reasoning to improve and enhance the services delivered by smart watches used to monitor heart rates. Keywords—elderly people; heart rate; Kingdom of Saudi Arabia; knowledge-based system

I.

I NTRODUCTION

Heart rate (HR) or pulse rate represents the number of times a heart beats each minute. The determination of what a normal heart rate is depends on many factors, such as age, body size, movement, exercise, and heart conditions. A normal heart rate can be between 70 and 100 beats per minute (bpm) [1]. In the United States (US), by 2030, it is projected that more than 8 million people will be diagnosed with heart failure (HF). The cost of HR treatment in 2012 was $31 billion, and it is estimated to reach $70 billion in 2030. The cost of treatment for HR in people who are 65 years

Russell J Campion2 Faculty of Computing, Engineering and Sciences 2 Staffordshire University Beaconside, Stafford, UK ST18 0AD +44 (0) 1785353464 [email protected]

of age and older is expected to increase from 69% in 2012 to 80% in 2030 [2]. In the US in 2009, 75% of patients who were hospitalised for HF were 65 years and older, and the cost of treatment was $37.2 billion [3]. In the United Kingdom (UK), cardiovascular disease (CVD) is the second leading cause of death, accounting for 28% of fatalities in 2014. The cost of treating CVD in England was more than £6.8 billion in 2012 [4]. The United Nations defines elderly people as people aged 60 years and older [5]. In 2015, the ageing population in the Kingdom of Saudi Arabia (KSA) was 5%. It is estimated that the ageing population will be at 20.9% in 2050. Furthermore, by 2100, this number is expected to increase dramatically to 33.5% [6]. In Saudi Arabia, roughly 20.7% of elderly people show up at primary health-care (PHC) facilities, and many fail to receive a proper diagnosis [7]. On some occasions, this leaves elderly people unable to obtain placement in beds at hospitals [8]. HF is a major concern for elderly people in the KSA [9], [10]. Elderly people often experience health issues that differ from those that they may have encountered when younger. As they age, they may need support in their daily activities and monitoring of their health [11]–[13]. Thus, the demographic issues related to elderly people and HF requires a suitable device for monitoring heart rate in elderly people. This paper is organised as follows: Section II describes smart watches, while Section III discusses its methodology. Section IV identifies the complementary approach, and finally, the conclusion and directions for future work are presented in Section V.

978-1-5090-4320-0/16/$31.00 ©2016 IEEE

II.

SMART W ATCH

A. Motivation The motivation for this research arose because some smart devices use accelerometers to monitor elderly people. However, if a failure occurs when an elderly person is watching TV, this will not be recognised by the device. A question was raised regarding the potential for determining whether an elderly person is alive or dead when he/she is watching TV in a home equipped with a variety of sensors, such as emotion, blood pressure, and temperature sensors. B. Definition Smart watches are computing devices whose operating systems offer limited capabilities [14], [15]. Smart watches have been used to monitor heart rate [16] through the use of sensors based on light technology that measure the rate of blood flow, as controlled by the pumping action of the heart [17].

Fig. 1: An Overview of Components and Process

C. Preferences

the mobile application installed on a smartphone or computer. This assists in synchronising the data from the smart watch to the mobile app. The smartphone uses a wireless connection to transfer data from the mobile app to the web portal. Fig. 1 provides an overview of the components and process.

The major aim of this study was to provide a technology that elderly people in Saudi Arabia could use to monitor their heartbeat and thus manage issues associated with HF. Results of a questionnaire distributed in the KSA demonstrated that elderly people in the country prefer wearable devices [18]. Therefore, smart watches were used in the experiments conducted.

B. Mobile App The mobile app is an application built by the smart watch provider to offer various services. In this study, the mobile app was installed on the elderly person’s smartphone. A user name and password were created on a login screen upon receipt of the device.

D. Problem Many smart watches are currently available on the market. We investigated the features of these smart watches in terms of the monitoring of heart rate. We found that most use applications and websites to visualise the heart rate data of the elderly person being monitored, and no specific actions can be taken if an abnormal heart rate occurs. E. Objective The objective of this paper is therefore to propose a complementary approach for smart watches that will provide a proactive response through the use of a knowledge-based system with rule-based reasoning. We used two smart watches (Apple Watch and Fitbit) to conduct the study and propose a complementary approach. III.

M ETHODOLOGY

A. System Requirements and Components Two concerns were considered important in choosing the smart watches: the presence of heartbeat sensors and compatibility with other devices. The Internet is the basis for use of the device. A Bluetooth network is used to connect the smart watch with

C. Online Portal The online portal is a website that allows users to visualise all activities and data whenever needed. For the current study, we focused on data related to heart rate. The data were uploaded to the online portal via the mobile app. D. Experimental Case An elderly person over the age of 60 was given a smart watch and monitored in two town locations and during his normal activities, including when he was driving. Ethical approval was granted before the experiment began. The elderly person was first monitored for seven days from February 18–24, 2016, in Stafford and Chepstow in the United Kingdom. Fig. 2 presents the 24-hour maximum and minimum heart rates for the seven days. The maximum rate was 149 bpm, and the minimum rate was 65 bpm. February 18, 19, and 20 displayed abnormal figures, as shown in Fig. 3. Thus, we contacted the elderly person to determine the reasons and learned that the abnormal readings were due to the person exercising, causing an elevated heart rate. The minimum rate was roughly as described in the literature review.

145

149

197

146 118

115

200

122

105

67

65

76

73

77

71

68

Max Min

Hear Rate

Heart Rate

160 140 120 100 80 60 40 20 0

150

120 104

115

100 56

59

58

59

112 63

Max Min

50 18

19

20

21

22

23

24

Days Fig. 2: Heart Rate for Elderly Person (18-24 February 2016)

0 20

21

22

23

24

Days Fig. 4: Heart Rate for Elderly Person (20 -24 May 2016)

Fig. 5: Heart Rate for Elderly Person (24th May 2016)

occurrence. We contacted the elderly person immediately after noticing the high pulse and learned that the elderly person was rushing to meet a deadline. However, the heart rate was within the normal range on all other days (Fig. 4). The elderly person was monitored using a web portal via a laptop or smartphone from three different countries: 1) Stafford, United Kingdom, 2) Jeddah, Saudi Arabia, and 3) Los Angeles, United States. The experimental process is illustrated in Fig. 6

Fig. 3: 24-Hour Heart Rate Data (18- 20 February 2016)

The elderly person was then monitored for five days from May 20–24, 2016 in the towns of Stafford and Chepstow in the United Kingdom. Fig. 4 presents the daily maximum and minimum heart rates for five days, from 7:30 a.m. to 11 p.m. The maximum rate was 197 bpm, and the minimum rate was 58 bpm. On May 21, 2016, the heart rate was 197 bpm, as shown in Fig. 5, which is quite high, indicting an abnormal

However, during the monitoring, the results show that we were only able to visualise the heartbeat data. The system currently does not take action when an abnormal even occurs. Helpful actions could include sending an alert message to family, relatives or health-care providers; dispatching a doctor or nurse; or dispatching an ambulance. Clearly, these kinds of systems require families, relatives and health-care providers to constantly monitor the elderly person’s heart rate, which can be extremely time consuming. In practice, smart watches are not linked to any support system of health-care providers and are mainly used to display visual data to the individual.

Fig. 6: The Process of Monitoring An Elderly Person

Therefore, we believe we can further develop these systems by adding features that will assist elderly persons with being healthy, independent and possibly living longer in their desired environment. Thus, we propose an approach to advance the current monitoring applications and systems, most of which only present the data.

E. Knowledge-based System A knowledge-based system (KBS) works by transferring knowledge for use in resolving issues and problems [19]. A KBS can be used to generate an intelligent action within a smart watch. Here, a KBS will be used along with rule-based reasoning, which is important for this approach. Rule-based reasoning is responsible for generating solutions based on the data [20], [21] acquired from the smart watch. The KBS acts as an expert system that can make appropriate decisions and take action when an abnormal event occurs. A set of rules will be developed according to the results from individual questionnaires regarding respondents' normal daily activities, the outcomes of the Community of Practice (CoP) and the literature. “Communities of Practice are groups of people who share a concern or a passion for something they do and learn how to do it better as they interact regularly” [22]. This concept has been outlined in previous work in which details of the model are presented [23]. IV. COMPLEMENTARY A PPROACH The proposed approach is based on the results of the experiment, with the goal of giving the elderly person an appropriate smart watch that can meet their needs. The experimental results show that curently, no intelligent actions can be taken when abnormal events occur. Therefore, a KBS should be built as an expert system that can take appropriate actions when abnormal

conditions occur. Rule-based reasoning is the most common option for creating a KBS. It can be represented through ‘if…then’ condition rules to determine the current status and assign appropriate solutions. An example of a rule-based system is illustrated in Fig. 7. Rule-based software will be purchased and modified according to the requirements defined by the CoP, the results of the elderly individuals' questionnaire based on their daily activities and the analysis of focus groups. The expert system will determine exceptions and issues related to elderly persons and will execute and run a set of rules to decide the appropriate action to resolve the issue and activate a proactive system if required. The smart watch collects the data from the elderly person and records them in a database system. Each person has a profile associated with his/her personal information stored in the database. Therefore, if the heart rate is less than 60 bpm or more than 100 bpm, the system may check the person’s profile and apply the following procedures or rules: Firstly, the system should try to contact the elderly person by telephone/text. If no response is received after a specified period, then: 1. if approval has been previously given to turn a camera on, a. the system will check the person’s health; if the person is ok, it will turn the camera off and store the data; or 2. if approval not been given to operate a camera, a decision is made to: a. telephone/text the elderly person again: i. if the elderly person responds satisfactorily, the system will store the data; or ii. if elderly person does not respond, then the system will text or call the family.

Experimentation using ZigBee technology for monitoring the movements of elderly people will also be undertaken to integrate smart watches with collected data from personalised daily activity questionnaires to identify the subjects’ usual activity patterns. Case-based reasoning will be investigated as an option for building the KBS and will be compared with the rulebased reasoning to determine the appropriate choice. References [1] [2]

[3]

[4]

[5]

British Heart Foundation, ‘Your heart rate’, 2014. P. A. Heidenreich, N. M. Albert, L. A. Allen, D. A. Bluemke, J. Butler, C. Fonarow, J. S. Ikonomidis, O. Khavjou, M. A. Konstam, T. M. Maddox, G. Nichol, M. Pham, I. L. Piña, and J. G. Trogdon, ‘Forecasting the Impact of Heart Failure in the United States’, Circ Heart Fail, vol. 6, no. 3, pp. 606–619, 2013. R. V. Samala, V. Navas, E. Saluke, and J. O. Ciocon, ‘Heart failure in frail, older patients: We can do “MORE”’, Cleveland Clinic Journal of Medicine, vol. 78, no. 12, pp. 837–845, 2011. N. Townsend, J. Williams, P. Bhatnagar, K. Wickramasinghe, and M. Rayner, Cardiovascular disease statistics, 2014. British Heart Foundation, 2014. United Nations, Department of Economic and Social Affairs, Population Division (2013). World Population Ageing 2013 Report. [Online]. 2013. United Nations publication. Available from: http://www.un.org/en/development/desa/population/publicatio ns/pdf/ageing/WorldPopulationAgeing2013.pdf.

[6]

[7] Fig. 7: Proposed Approach [8]

V.

CONCLUSION AND FUTURE WORK

In this research, we have used two different types of smart watches to monitor an elderly person in different locations. The elderly person was monitored from three different countries. The results show that the systems currently installed and incorporated into smart watches are only meant for visualisation; no intelligent action can be taken when abnormal events occur. Therefore, we present a complementary approach for these systems by proposing a knowledge-based system combined with a rule-based system.

[9]

[10]

[11]

[12]

Future work will include conducting a CoP with Age UK to gain knowledge about their system. The goal of this will be to share information and experiences with practitioners. This technique will help researchers transfer the knowledge from Age UK to the stakeholders in the KSA. This will be accomplished by holding meetings and documenting the mission, objectives, and processes.

[13] [14]

[Accessed:

12 January 2016]. United Nations, Department of Economic and Social Affairs, Population Division (2015). WorldPopulation Prospects: The 2015 Revision, Key Findings and Advance Tables. Working Paper No. ESA/P/WP.241. M. H. Al-Doghether, ‘Prescribing in primary care for the older people’, Saudi Medical Journal, vol. 25 (4), no. December, pp. 488–492, 2004. MOH (22013). Facts Achievements Report. [Online]. 22013. Minsitry of Health. Available from: http://www.moh.gov.sa/Ministry/About/Documents/NEW1434-1430.pdf. [Accessed: 12 January 2016]. M. AbuRuz, F. Alaloul, A. Saifan, R. Masa’deh, and S. Abusalem, ‘Quality of Life for Saudi Patients With Heart Failure: A Cross-Sectional Correlational Study’, Global Journal of Health Science, vol. 8, no. 3, pp. 49–58, 2015. M. Albugami, Y. Al Twaijri, H. Janaid, A. Heggi, A. Elamin, R. Abdelreheem, S. Mohammed, W. Alahmadi, and N. Alturaif, ‘Health Care Maintenance of Saudi Elderly Women in Outpatient Clinics’, American Journal of Internal Medicine, vol. 3, no. 5, p. 197, 2015. A. Hossain, A. Alamri, A. Almogren, K. A. Hossain, and J. Parra, ‘A framework for a context-aware elderly entertainment support system’, Sensors (Basel, Switzerland), vol. 14, no. 6, pp. 10538–10561, 2014. A. Al Saif, E. Waly, and S. Alsenany, ‘The Prediction of Falls Among older people in Saudi Arabia’, Journal of American Science 2012, vol. 8, no. 6, 2012. R. Blasco, Á. Marco, R. Casas, D. Cirujano, and R. Picking, ‘A smart kitchen for ambient assisted living.’, Sensors (Basel, Switzerland), vol. 14, no. 1, pp. 1629–1653, 2014. C. Xu and K. Lyons, ‘Shimmering Smartwatches: Exploring the Smartwatch Design Space’, in Proceedings of the Ninth International Conference on Tangible, Embedded, and Embodied Interaction - TEI ’14, Stanford, California, USA,

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22] [23]

2015, pp. 69–76. K. Lyons, ‘What can a dumb watch teach a smartwatch?’, in Proceedings of the 2015 ACM International Symposium on Wearable Computers - ISWC ’15, Osaka, Japan., 2015, pp. 3– 10. Z. Zhang, ‘Photoplethysmography-based heart rate monitoring in physical activities via joint sparse spectrum reconstruction’, IEEE Transactions on Biomedical Engineering, Texas, USA, vol. 62, no. 8, pp. 1902–1910, 2015. S. Rios-Aguilar, J. L. M. Merino, A. Millán Sánchez, and Á. Sánchez Valdivieso, ‘Variation of the Heartbeat and Activity as an Indicator of Drowsiness at the Wheel Using a Smartwatch’, International Journal of Interactive Multimedia and Artificial Intelligence, vol. 3, no. 3, p. 96, 2015. M. H. Alsulami and A. S. Atkins, ‘Elderly Saudi Arabians’ Perceptions and Attitudes Towards using Ambient Assisted Living Technologies’, Proceedings of 64th The IIER International Conference, Barcelona, Spain, 4th March 2016, ISBN: 978-93-85973-53-6, no. March, 2016. K. P. Tripathi, ‘A Review on Knowledge-based Expert System : Concept and Architecture’, Artificial Intelligence Techniques - Novel Approaches & Practical Applications, vol. 4, no. 4, pp. 19–23, 2011. I. Bichindaritz, E. Kansu, K. M. Sullivan, and F. Hutchinson, ‘Integrating Case-Based Reasoning,Rule-Based Reasoning and Intelligent Information Retrieval for MedicalProblemSolvin’, Information Retrieval, pp. 22–27, 1998. S. L. Syed-abdullah, H. Daniyal, R. Rashid, S. Noor, and N. Sharif, ‘Rule Based Reasoning and Case Based Reasoning Techniques for Juvenile Delinquency Legal Reasoning Model’, in Proceedings of the 5th International Conference on Computing and Informatics, ICOCI 2015, Istanbul, Turkey, 2015, no. 192, pp. 677–683. E. Wenger and B. Trayner-Wenger, ‘Communities of practice: a brief introduction’, in April 2015, 2...