Towards a Fully Automated Bracelet for Health Emergency Solution

Igor Matias

1,2

, Nuno Pombo

1,2,3

and Nuno Garcia

1,2,3

Computer Science Department, Universidade da Beira Interior,

R. Marquês de Ávila e Bolama, 6201-001, Covilhã, Portugal

Instituto de Telecomunicações, Av. Rovisco Pais, 1, 1049-001, Lisboa, Portugal

Universidade Lusófona de Humanidades e Tecnologias, Campo Grande, 376, 1749-024, Lisboa, Portugal

Keywords: Bio Signal Monitoring Device, Arduino, Sensors, RF Communications, Emergency Scenarios Detection.

Abstract: Activity and health monitoring bracelets are currently a popular consumer electronic wearable, and several

different manufacturers market several different versions, integrating a wide and diverse range of sensors.

The several different types of bracelets often require the user to interact with the bracelet itself or with the

help line attendant. However, this interaction is not suitable in many emergency scenarios. Thus, this paper

presents a novel system to monitor, detect and communicate health critical situations, in a full automated

manner. The system encompasses a bracelet capable for health abnormalities detection based on both vital

signs, and accelerometer data collection from the user. This paper also describes the design and prototyping

process of the bracelet, providing insight and solutions to observed problems. In spite of the early stage of

this system, the observed results are promising and offer room for improvements on either complementary

studies or different clinical emergence scenarios.

1 INTRODUCTION

According to (United Nations, 2015), by the year of

2015, Europe, Oceania and North America, had

about 20 percent of elderly people (60 or more years

old). By 2050, these numbers are expected to

become higher than 30 percent. Authors of

(Administration on Aging; Administration for

Community Living; U.S. Department of Health and

Human Services, 2012) indicate that, only in the

United States of America, by 2011, the number of

persons aged 65 years old or more was representing

41.4% of the total country’s population, number

expected to increase to about 56% in 2020. The

authors also indicate that “about 28% (11.8 millions)

of noninstitutionalized older persons live alone (8.4

million women, 3.5 million men)”, where “almost

half of older women (46%) age 75+ live alone”.

As studied in (“Saúde Pública - Fatores

associados a quedas em uma coorte de idosos

residentes na comunidade Fatores associados a

quedas em uma coorte de idosos residentes na

comunidade,” n.d.), the fact that an elderly is living

alone contributes to the increase of the chance of

occurring a fall, representing 28,5 percent in 1667

studied Brazilian elderly people, during two years.

Living alone, described as “loneliness” in (Barg et

al., 2006), is even “highly associated with depressive

symptoms, anxiety, and hopelessness”.

Depression being so present among people with

more than 60 years old and taking into account that

“high levels of depressive symptomatology are

associated with increased risks of myocardial

infarction (MI) and mortality” ((Barefoot and

Schroll, 1996)), elderly living alone is a current and

an expected future scenario for many families,

which raises pertinent challenges due to the fact that

accidents may occur with elderly and frail people,

which may result in morbidity or eventually in

death.

Only in the United States of America (USA), as

presented in (Lloyd-Jones et al., 2010), from the

total amount of deaths caused by Cardiovascular

Disease (CVD) in all ages during 2006, the origin

for 51 percent of the cases was Coronary Heart

Disease (CHD) and Stroke for 17 percent. These

same data show that approximately 677 thousand

deaths were recorded due to CVD in people aged 65

years or older, and were also noted approximately 33

thousand deaths caused by Heart Failure (HF) in

people 60 years of age or older, in the same year as

the previous.

Matias, I., Pombo, N. and Garcia, N.

Towards a Fully Automated Bracelet for Health Emergency Solution.

DOI: 10.5220/0006800803070314

In Proceedings of the 3rd International Conference on Internet of Things, Big Data and Security (IoTBDS 2018), pages 307-314

ISBN: 978-989-758-296-7

307

Notwithstanding every emergency solution is

intended to be the fastest and easiest possible, the

current solutions are merely based on emergency

calls, wrist bracelets, or pendant devices with an

emergency button. Therefore, its application in real

life is limited to cases when the user is conscious

and at the same time is susceptible to limitations due

to the “human factor”, in either the call duration or

the difficult to press the button.

Although every created solution tried to decrease

the time until the receipt for therapeutic strategies,

delays such as “human factor”, manual emergency

buttons, people avoiding ambulance services when

they really need it and others, “prevent the early

application of life-saving procedures and contribute

substantially to a diminished effectiveness of

treatment” ((Luepker RV et al., 2000)).

Thus, this paper describes the design and the

implementation of a fully automated device capable

of recognizing several types of health conditions, as

well as contacting the emergency services when

needed, using the vital signals collected from the

users’ wrist.

The proposed system also provides a personal

database of collected vital signals, which will then

be used by a doctor to detect any abnormalities in an

early stage, also as providing other ways to decrease

the emergency response time.

In summary, the main contribution of this study

is to define a fully automated system architecture to

measure vital signals, in order to detect movement

abnormalities or hearth related diseases and,

therefore, to alert physicians and/or informal health

care providers on alarm episodes, while using

mostly open access electronics. The ultimate goal is

to provide an intelligent and autonomous solution to

help persons living alone or being monitored due to

a chronical disease.

2 BACKGROUND

As presented in Table 1, there are already on the

market several solutions for the simple “emergency

button” option, such as (Bay Alarm Medical, n.d.;

Rescue Alert, n.d.; LifeStation Inc., n.d.; Alert1,

n.d.; Philips N.V., n.d.; ADT LLC, n.d.; Life Fone,

n.d.; Medical Guardian LLC, n.d.; Solodev, n.d.;

Samsung, n.d.; FitBit, n.d.; Mi, n.d.), some having

an extra feature like Global Position System tracking

or fall-detection and others being able to collect

some important health data used in this new

approach, such as heart rate and body temperature

(smart bands and smart watches).

However, none offers a way to interlink the

collected data to detect health conditions and/or

immediate emergency situations, this is, some of the

current solutions already have some of the same

sensors this new approach uses, but they are not

using its data to recognize health

conditions/emergencies.

For example, by using a pulse monitor (capable

of reading heart rate variability and others), it is

possible to detect arrhythmia ((Patel et al., 2012;

Tsipouras and Fotiadis, 2004)), CVD ((“HEART

RATE AS A PROGNOSTIC FACTOR FOR

CORONARY HEART DISEASE AND

MORTALITY: FINDINGS IN THREE CHICAGO

EPIDEMIOLOGIC STUDIES | American Journal of

Epidemiology | Oxford Academic,” n.d.)), Coronary

Artery Disease (CAD) ((Kligfield et al., 1989)), HF

((Nolan et al., 1998)) or even a stroke ((Glotzer et

al., 2003; Ritter et al., 2011)). Using a body

temperature sensor, it is still possible to establish a

relation between hypothermia/hyperthermia and

longevity after a stroke happen ((Kammersgaard et

al., 2002)). Finally, by using an accelerometer to

monitor the user’s sleep, it’s possible to detect

certain episodes and/or disorders, described in

(Hjorth et al., 2012).

Thus, to decrease the lack of assistance in

emergency situations, it is necessary to create a

device capable of, using mainly already existing

sensors, detect and automatically react to health

abnormalities, being anyway portable and easy to

use.

Table 1: List of Features of Each Device.

Features

Name of solution

Fall detection/accelerometer

GPS locator

Heart rate monitor

Body temperature

Skin detection

Health record

Emergency button/contact

Bay Alarm

Medical ((Bay

Alarm Medical,

n.d.))

Rescue Alert

((Rescue Alert,

n.d.))

IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security

308

Table 1: List of Features of Each Device. (cont.)

LifeStation

((LifeStation Inc.,

n.d.))

Alert1 ((Alert1,

n.d., p. 1))

Philips LifeLine

((Philips N.V.,

n.d.))

ADT Medical

Alert ((ADT LLC,

n.d.))

LifeFone ((Life

Fone, n.d.))

Medical

Guardian

((Medical

Guardian LLC,

n.d.))

MobileHelp

((Solodev, n.d.))

Samsung Gear S3

((Samsung, n.d.))

FitBit alta HR

((FitBit, n.d.))

Mi Band 2 ((Mi,

n.d.))

Proposed system

3 METHODS

This section describes the proposed system,

including its components, and decision workflow. In

this study, a prototyping platform, known as

Arduino, was used. In fact, two units were used, one

to collect data from the sensors and other to collect

data sent from the first, which is stored and

interpreted, combined with the following sensors:

 3-axis accelerometer ((Analog Devices Inc.,

2009));

 GPS antenna ((Sparkfun Eletronics, n.d.));

 Pulse sensor ((Sparkfun Electronics, n.d.));

 Body temperature sensor ((Sparkfun

Electronics, n.d.));

 Skin contact detection sensor;

 Waterproof speaker ((VISATON GmbH & Co.

KG, n.d.));

 434 MHz radio frequency transmitter and

receiver ((Sparkfun Electronics, n.d.));

All these sensors, excluding the skin contact

detection, are off-the-shelf components, and

available on the market, not from medical grade.

The skin contact detector was designed and built for

this specific study. It is composed by two electrical

pins (wires) and one 1 kilo ohm resistor (similar to

the human skin resistance, as studied and concluded

in (Ho et al., 2012)), as illustrated in Fig. 1, referring

the output power of the Arduino board.

Figure 1: Skin contact detection sensor scheme.

The data collector cable was connected to an

analogue input port, which receives 10 bits’ data,

and the electrical pins are designed to touch the skin

of the user (placed on his/her arm).

All the boards’ ports used to connect all the

sensors are listed in Table 2.

Table 2: Ports and Sensors.

Input/output

Sensor(s) connected

GND

Accelerometer, GPS antenna, Pulse

sensor, Body temperature sensor, RF

transmitter, RF receiver, Speaker, Skin

contact detection sensor

3.3V

Accelerometer

GPS antenna, Pulse sensor, Body

temperature sensor, RF transmitter, RF

receiver, Skin contact detection sensor

Accelerometer

Skin contact detection sensor

Pulse sensor

GPS antenna

Body temperature sensor

RF transmitter (part of the device for the

user to wear), RF receiver (part of the device

to collect data to store)

Speaker

This project is composed by two distinct parts,

assembled in two Arduino boards.

On the collector and transmitter board we

assembled all the sensors to collect data, the speaker

and the RF transmitter. This part was responsible to

be on the users’ wrist, powered by a 7.0 volts’

Towards a Fully Automated Bracelet for Health Emergency Solution

309

battery, to read all the sensors and send the

information collected, each 10 seconds, to the other

board.

On the other hand, the receiver and interpreter

board was connected to a computer and, by reading

the data received from the RF receiver, checked if

any emergency scenario (described below) were

detected and, if necessary, sounded an alarm,

contacting the emergency services automatically,

also sending the recorded vital signals data of the

last 24 hours.

By combining these two concepts and running

the scenarios described below, the device can

produce alarms, that can be deactivated in case of a

false alarm, establish the automated contact with the

emergency services, send sensors data from the

moment and the last hours to that crew and guide

them to the victim, using sound.

Some of the already designed scenarios need to

be improved and tested in real cases, being then

interpreted by an Artificial Intelligence.

Besides very basic yet, they are the following:

 Detect if the device is placed on the user’s

wrist:

o If the data collected from the skin contact

detection sensor is higher than “200”

(skin detected), proceed to the next

scenarios. Else, repeat this scenario.

 Pulse abnormalities detection:

o If the pulse is below 30 beats per minute

(BPM) or higher than 140 BPM, check

temperature;

o If the body temperature is under 38ºC and

higher than 36ºC, check the pulse during

the next 20 seconds;

o If, during the next 20 seconds, the pulse

average is lower than 35 BPM or higher

than 135 BPM, start “emergency

procedure”;

o (the BPM and body temperature values

will be adapted to each user of the

prototype, regarding his/her previous

recorded data and known health

condition);

 Body temperature abnormalities detection:

o If the body temperature is lower than

36ºC or higher than 39ºC, repeat this

scenario during the next 30 seconds;

o If in the next 30 seconds the temperature

average is lower than 34ºC or higher than

40ºC, start “emergency procedure”.

 Movement abnormalities detection (falls or

abrupt movements originated by some type of

injury):

o If there are “G-force” readings higher

than 3.55 G (fall peaks studied in (Bourke

et al., 2007)), check data from the

accelerometer during the next 10 seconds;

o If, in the next seconds, another positive of

the first point is detected or the “G-force”

average is lower than 1.2 G, start

“emergency procedure” (it was detected a

fall and the user isn’t moving anymore);

o (the movement values will be studied

more closely in the next phase of this

research).

 Emergency procedure:

o The alarm is sounded by the speaker on

the collector and transmitter device;

o If, during the next 30 seconds, a “false

alarm” button in the collector device is

not pressed (to implement on the next

phase of this research), start the

automated call to the emergency services,

sending the location of the person

(detected by the GPS antenna) and the

data collected 30 seconds before the

scenario was detected;

o When it’s confirmed that an emergency

crew is on the way to the user’s location

to help, start sending the data collected

every 30 seconds, preceded by the

sending of the records of the last 24

hours.

Moreover, all the collected data are stored during 24

hours on the collector and then sent to a remote

database, which can be accessed and used by a

physician to detect some health issue not causing an

emergency.

4 RESULTS

By assembling the two concept parts, it was possible

to collect and interpret the “normal” vital signals

data from a person on a non-controlled environment

or state.

Making an automated device capable of

detecting health issues means it needs to have

sensors capable of measuring certain vital signals. In

line with this, the presented project is designed to be

able to recognize patient’s behavioural abnormality,

based on heartbeat, body temperature and body

movements.

IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security

310

Related to the heart, this type of device would

have to be capable of detecting, for example, a

stroke. According to (Ritter et al., 2011), a stroke

can be diagnosed and evaluated using the pulse

BPM collected data and, as studied in (Glotzer et al.,

2003), it can also be detected thru Atrial High Rate.

Other important heart disease is arrhythmia which,

as tested in (Patel et al., 2012), is possible to detect

through the “R-R interval” (also known as “Inter

Beat Interval”). With Heart Rate Variability, it is

even possible to detect a Heart Failure case ((Nolan

et al., 1998)) and, as concluded in (Kligfield et al.,

1989), a Coronary Artery Disease can also be

assumed using data collected from heart rate.

Finally, it is also possible to establish a connection

between the heart rate and the probability of having

Cardiovascular Diseases, thus being an important

way of diagnose. To solve and be able to detect

these five studied types of heart diseases we used a

pulse sensor (Sparkfun Electronics, n.d.), measuring

BPM, ST segment and IBI of the patient.

About body temperature, it is needed to measure

the users’ one, allowing the device to detect any

abnormalities as described in (Lu and Dai, 2009)

(more specific to people above 65 years old) and

recognize patterns for diagnose after a stroke has

occur, as concluded in (Kammersgaard et al., 2002).

For that, the solution was using a body temperature

sensor found in (Sparkfun Electronics, n.d.).

Finally, using a 3-axis accelerometer ((Analog

Devices Inc., 2009)) the device can detect falls,

false-alarms and sleep monitoring the user, as

detailed in (Hjorth et al., 2012).

The only scenario coded so far was a small part

of the movement abnormalities detection one.

Because there were only tested the average signals,

it is not possible to present any significant results of

this at this stage of the project.

The Fig. 4 shows the current prototype of the

receiver and interpreter board, as well as the current

prototype of the collector and transmitter

(“Intelligent Health Monitoring Bracelet” – adapted

to a real-life bracelet – Fig.5). The current prototype

of the collector is yet in a large scale, which will be

reduced to a simple and discrete bracelet, replacing

the Arduino board with a smaller version for

wearable prototypes, also removing the extra amount

of wires and replacing some sensors with other

smaller and equivalent versions of them.

Figure 2: Current prototype of the receiver and interpreter.

Figure 3: Current prototype of the collector and

transmitter.

After assembling these two parts of the device,

we found some issues on working with all the

sensors together:

 Receiver and interpreter:

o The length of the antenna had to be

calculated depending on the frequency of

the transmissions (0,17 metres as

calculated in (1)), the same that was

applied to the collector and transmitter. If

lower or higher than 17 centimetres, the

communication efficiency was lower in

distance (the average was around 15

metres).









≈ 0,691244

(1)

¼ of antenna wavelength =





≈ 0,17 metres

 Collector and transmitter:

o The GPS antenna had to be tested

separately to the remaining sensors,

because when all were plugged to the

“5V” port, the GPS antenna didn’t receive

the minimal amount of current and

wouldn’t start seeking a satellite signal;

o The accelerometer sometimes presented

some values over 7 G-forces, due to bad

connections of the sensors’ pins;

Towards a Fully Automated Bracelet for Health Emergency Solution

311

The skin contact detection sensor works best in non-

fully dry skin, i.e. with some humidity in the human

tissue (because the resistance is lower – more similar

to the resistor used). To solve this, we forced the

sensor against the user skin, working as expected.

5 DISCUSSION AND

CONCLUSIONS

By crossing data from three sensors, namely, GPS,

accelerometer, pulse sensor, body temperature and

skin contact detection sensor, on the presented

scenarios and implementing the two devices

communication system, it is now possible to create a

new e-health and intelligent emergency solution,

with new features, complementing the ones

described in the Introduction section. The proposed

solution may solve several limitations observed on

the literature:

 By removing the “human factor” during a routine

or an emergency call this is a quicker solution;

 By sending the collected data from the 30

seconds before an emergency scenario detection

and every 30 seconds after, it allows the medical

crew to have real-time vital signals from the

user, decreasing the “first medical exam”

duration, done when arriving to the patients’

home, also decreasing the helping time,

increasing the chances of a good recovery of the

person;

 By having an intelligent and advanced

emergency scenarios software detector, it is

avoided having an “emergency button” present

on the current devices, this way reducing the

chance of a “false alarm” or a false press of the

button;

 With the data being stored and sent to a database,

it allows a doctor to constantly remotely monitor

the health of the user, being possible to detect

several health issues in an early stage.

The skin contact detection sensor had to be

implemented due to the non-exact values readied

from the pulse sensor (when not placed on the user’s

finger the values derivate from below 20 BPM and

higher than 200 BPM often) – this sensor has to be

on the users’ finger because it was designed for that

and when tested on the arm the values were not

correct (average of 20 BPM of difference to the real

ones). This same sensor was built by us, because we

didn’t find any identical on the market.

After testing the reading process of the sensors

data, we realize the storage capability of the Arduino

board was not enough and preferred to send them to

the receiver and interpreter device, being then stored

in a computer. The same was applied to the

scenarios recognition, offering a decrease of power

consume of the battery.

Regarding the frequency on which the data must

be collected, it is needed to study more deeply the

ideal of collections in one minute (the minimum that

won’t affect the emergency solution efficiency), due

to the high power consumption of the current one,

which is constantly reading new values.

For the communicators antenna length, it will be

sought a better antenna architecture, being then

possible to implement it more easily on a real life

wrist bracelet.

The list of the three sensors used were chosen

due to the data collected and described in the Results

section.

Related to the radio frequency communicators,

they were chosen instead of “Wi-Fi” or “Bluetooth”

due to the lower power consumption.

6 FUTURE WORK

Because this is an early stage of this project, it’s

only possible to us to present the work done so far.

On its next stage, there will be presented:

 The results of the tests, involving the

scenarios detection (improve of the

current ones and addition of new ones);

 The result of implementing an Artificial

Intelligence algorithm for analysing the

health data;

 The power consumption of each sensor

used;

 The battery power management

efficiency (turning on and off the

sensors to economize it);

 Detailed success rates for the current

communication process and its

efficiency affected by distance, amount

of data and environment;

 Possible communications alternatives,

to prevent data loss when in rooms with

many objects;

 A new and more easy to use version of

the bracelet, using smaller and more

efficient sensors;

 The results of the tests of the overall

system (synchronization, success rate of

IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security

312

the communications and alarm

generation and success rate for the

health issue recognition).

ACKNOWLEDGEMENTS

Contributing to this research, the authors affiliated

with the Instituto de Telecomunicações also

acknowledge the funding for the research by means

of the program FCT project UID/EEA/50008/2013.

The authors would also like to acknowledge the

contribution of the COST Action IC1303 –

AAPELE – Architectures, Algorithms and Protocols

for Enhanced Living Environments.

REFERENCES

Administration on Aging; Administration for Community

Living; U.S. Department of Health and Human

Services, 2012. A Profile of Older Americans: 2012.

ADT LLC, n.d. Medical Alert System & Fall Detection

Monitoring from ADT [WWW Document]. URL

https://www.adt.com/health (accessed 9.29.16).

Alert1, n.d. Alert1 Medical Alert Services for Seniors

[WWW Document]. URL https://www.alert-1.com

(accessed 9.29.16).

Analog Devices Inc., 2009. ADXL335 3-Axis

Accelerometer Module [WWW Document]. URL

http://www.robotshop.com/en/3-axis-accelerometer-

module.html (accessed 2.26.17).

Barefoot, J.C., Schroll, M., 1996. Symptoms of

Depression, Acute Myocardial Infarction, and Total

Mortality in a Community Sample. Circulation 93,

1976. https://doi.org/10.1161/01.CIR.93.11.1976

Barg, F.K., Huss-Ashmore, R., Wittink, M.N., Murray,

G.F., Bogner, H.R., Gallo, J.J., 2006. A Mixed-

Methods Approach to Understanding Loneliness and

Depression in Older Adults. J. Gerontol. Ser. B 61,

S329–S339. https://doi.org/10.1093/geronb/61.6.S329

Bay Alarm Medical, n.d. Bay Alarm Medical: Medical

Alert Systems for Seniors [WWW Document]. URL

http://www.bayalarmmedical.com (accessed 9.29.16).

Bourke, A.K., O’Brien, J.V., Lyons, G.M., 2007.

Evaluation of a threshold-based tri-axial accelerometer

fall detection algorithm. Gait Posture 26, 194–199.

https://doi.org/10.1016/j.gaitpost.2006.09.012

FitBit, n.d. FitBit alta HR [WWW Document]. URL

https://www.fitbit.com/altahr (accessed 4.20.17).

Glotzer, T.V., Hellkamp, A.S., Zimmerman, J., Sweeney,

M.O., Yee, R., Marinchak, R., Cook, J., Paraschos, A.,

Love, J., Radoslovich, G., Lee, K.L., Lamas, G.A.,

2003. Atrial High Rate Episodes Detected by

Pacemaker Diagnostics Predict Death and Stroke.

Circulation 107, 1614. https://doi.org/10.1161/01.

CIR.0000057981.70380.45

HEART RATE AS A PROGNOSTIC FACTOR FOR

CORONARY HEART DISEASE AND

MORTALITY: FINDINGS IN THREE CHICAGO

EPIDEMIOLOGIC STUDIES | American Journal of

Epidemiology | Oxford Academic [WWW Document],

n.d. URL https://academic.oup.com/aje/article-

abstract/112/6/736/133827 (accessed 1.26.18).

Hjorth, M.F., Chaput, J.-P., Damsgaard, C.T., Dalskov, S.-

M., Michaelsen, K.F., Tetens, I., Sjödin, A., 2012.

Measure of sleep and physical activity by a single

accelerometer: Can a waist-worn Actigraph adequately

measure sleep in children? Sleep Biol. Rhythms 10,

328–335. https://doi.org/10.1111/j.1479-8425.2012.

00578.x

Ho, J.D., Dawes, D.M., Kroll, M.W., 2012. Atlas of

Conducted Electrical Weapon Wounds and Forensic

Analysis, 1st ed. Springer-Verlag New York, New

York.

Kammersgaard, L.P., Jørgensen, H.S., Rungby, J.A.,

Reith, J., Nakayama, H., Weber, U.J., Houth, J.,

Olsen, T.S., 2002. Admission Body Temperature

Predicts Long-Term Mortality After Acute Stroke.

Stroke 33, 1759. https://doi.org/10.1161/01.STR.

0000019910.90280.F1

Kligfield, P., Ameisen, O., Okin, P.M., 1989. Heart rate

adjustment of ST segment depression for improved

detection of coronary artery disease. Circulation 79,

245. https://doi.org/10.1161/01.CIR.79.2.245

Life Fone, n.d. Medical Alert System & Medical Alarm |

LifeFone [WWW Document]. URL

https://www.lifefone.com (accessed 9.29.16).

LifeStation Inc., n.d. Medical Alert System & Medical

Alarms | LifeStation [WWW Document]. URL

https://www1.lifestation.com/ (accessed 9.29.16).

Lloyd-Jones, D., Adams, R.J., Brown, T.M., Carnethon,

M., Dai, S., De Simone, G., Ferguson, T.B., Ford, E.,

Furie, K., Gillespie, C., Go, A., Greenlund, K., Haase,

N., Hailpern, S., Ho, P.M., Howard, V., Kissela, B.,

Kittner, S., Lackland, D., Lisabeth, L., Marelli, A.,

McDermott, M.M., Meigs, J., Mozaffarian, D.,

Mussolino, M., Nichol, G., Roger, V.L., Rosamond,

W., Sacco, R., Sorlie, P., Stafford, R., Thom, T.,

Wasserthiel-Smoller, S., Wong, N.D., Wylie-Rosett,

J., 2010. Heart Disease and Stroke Statistics—2010

Update. Circulation 121, e46. https://doi.org/

10.1161/CIRCULATIONAHA.109.192667

Luepker RV, Raczynski JM, Osganian S, et al, 2000.

Effect of a community intervention on patient delay

and emergency medical service use in acute coronary

heart disease: The rapid early action for coronary

treatment (react) trial. JAMA 284, 60–67.

https://doi.org/10.1001/jama.284.1.60

Lu, S.-H., Dai, Y.-T., 2009. Normal body temperature and

the effects of age, sex, ambient temperature and body

mass index on normal oral temperature: A prospective,

comparative study. Int. J. Nurs. Stud. 46, 661–668.

https://doi.org/10.1016/j.ijnurstu.2008.11.006

Medical Guardian LLC, n.d. Medical Alert Systems,

Devices & Alarms for Seniors in Emergencies |

Towards a Fully Automated Bracelet for Health Emergency Solution

313

Medical Guardian [WWW Document]. URL

https://www.medicalguardian.com (accessed 9.29.16).

Mi, n.d. Mi Band 2 [WWW Document]. URL

http://www.mi.com/en/miband2/ (accessed 4.20.17).

Nolan, J., Batin, P.D., Andrews, R., Lindsay, S.J.,

Brooksby, P., Mullen, M., Baig, W., Flapan, A.D.,

Cowley, A., Prescott, R.J., Neilson, J.M.M., Fox,

K.A.A., 1998. Prospective Study of Heart Rate

Variability and Mortality in Chronic Heart Failure.

Circulation 98, 1510. https://doi.org/10.1161/01.CIR.

98.15.1510

Patel, A.M., Gakare, P.K., Cheeran, A.N., 2012. Real

Time ECG Feature Extraction and Arrhythmia

Detection on a Mobile Platform.

Philips N.V., n.d. How We Help [WWW Document].

URL https://www.lifeline.philips.com (accessed

9.29.16).

Rescue Alert, n.d. Medical Alert Systems | Rescue Alert

[WWW Document]. URL https://www.rescuealert.

com (accessed 2.28.17).

Ritter, M.A., Rohde, A., Heuschmann, P.U., Dziewas, R.,

Stypmann, J., Nabavi, D.G., Ringelstein, B.E., 2011.

Heart rate monitoring on the stroke unit. What does

heart beat tell about prognosis? An observational

study. BMC Neurol. 11, 47. https://doi.org/10.1186/

1471-2377-11-47

Samsung, n.d. Samsung Gear S3 [WWW Document].

URL http://www.samsung.com/us/explore/gear-s3/

#specs (accessed 4.20.17).

Saúde Pública - Fatores associados a quedas em uma

coorte de idosos residentes na comunidade Fatores

associados a quedas em uma coorte de idosos

residentes na comunidade [WWW Document], n.d.

URL https://scielosp.org/scielo.php?pid=S0034-

89102002000700008&script=sci_arttext#ModalArticl

es (accessed 1.27.18).

Solodev, n.d. Medical Alert Systems from MobileHelp

[WWW Document]. URL http://www.mobilehelp.com

(accessed 9.29.16).

Sparkfun Electronics, n.d. Pulse Sensor - SEN-11574 -

SparkFun Electronics [WWW Document]. URL

https://www.sparkfun.com/products/11574 (accessed

9.29.16a).

Sparkfun Electronics, n.d. Temperature Sensor -

Waterproof (DS18B20) - SEN-11050 - SparkFun

Electronics [WWW Document]. URL

https://www.sparkfun.com/products/11050

Sparkfun Electronics, n.d. RF Link Transmitter - 434MHz

- WRL-10534 - SparkFun Electronics [WWW

Document]. URL https://www.sparkfun.com/

products/10534 (accessed 9.29.16c).

Sparkfun Eletronics, n.d. GPS Receiver - EM-506 (48

Channel) - GPS-12751 - SparkFun Electronics.

[WWW Document]. URL https://www.sparkfun.com/

products/12751 (accessed 9.29.16).

Tsipouras, M.G., Fotiadis, D.I., 2004. Automatic

arrhythmia detection based on time and time–

frequency analysis of heart rate variability. Comput.

Methods Programs Biomed. 74, 95–108.

https://doi.org/10.1016/S0169-2607(03)00079-8

United Nations, 2015. World Population Ageing [WWW

Document]. URL http://www.un.org/en/development/

desa/population/publications/pdf/ageing/WPA2015_R

eport.pdf (accessed 2.27.17).

VISATON GmbH & Co. KG, n.d. K 36 WP - 8 Ohm

[WWW Document]. URL http://www.visaton.de/en/

industrie/klein/k36wp_8.html (accessed 9.29.16).

IoTBDS 2018 - 3rd International Conference on Internet of Things, Big Data and Security

314