SENSECARE

Real-time Location-based Health Monitoring System

Gerardo Reveriego, Javier Tellez, Cristina Urdiales and Francisco Sandoval

Departamento de Tecnologia Electronica, Universidad de Malaga

E.T.S.I. Telecomunicacion, Campus de Teatinos S/N, Malaga, Spain

Keywords:

Telehealth, Physiological parameters, Biotelemetry, GIS.

Abstract:

This paper presents an application capable of monitoring the activity of a user in real time. Our work relies

on a mobile phone since most users already have one and usually take it with them everywhere they go. Our

system makes use of a pulse oximeter connected to the phone to register physiological parameters and sends

them to a coordination centre. Other users, such as doctors or carers, can download health data from the

coordination center. The result is a report on the route followed by the user with geo-referenced physiological

parameters shown in real time. A Geographic Information System (GIS) was used as support. This standard-

based report can be visualized with various compatible GIS. We have used Google Earth due to its public and

free accessibility.

1 INTRODUCTION

It has been estimated that just in the USA, the health

care system could reduce costs by nearly $ 200 bil-

lion during the next 25 years if remote patient mon-

itoring tools were delivered for the most prevalent

diseases, such as congestive heart failure, COPD

(Chronic Obstructive Pulmonary Disease), diabetes,

chronic wounds or skin ulcers (Litan, 2008).

In the ﬁeld of telehealth, remote monitoring or

telemonitoring is a broad area that has usually in-

volved the deployment on the user’s home of one or

various kind of sensors that will measure speciﬁc vital

signs. Hitherto, many countries and health providers

have focused on applications, such as basic videocon-

ference tools or monitoring devices. Many of these

home-based telemonitoring solutions are basically a

deployment of a single or several medical devices:

spirometers, heart monitors, blood pressure cuffs and

so on. Commercial examples of home health mon-

itoring devices are: Health Buddy, Ideal LIFE Pod,

Genesis DM, Intel’s Health Guide, LifeView.

A small number of products can use a mobile de-

vice as a communication gateway, like Telestation or

HealthAnywhere. This feature, have been proposed

in (Pandian, 2008), (Romero et al., 2008), (Jones,

2006a), (Jones, 2006b). Our system comprises a wire-

less pulse oximeter and a mobile phone that collects

the data from the sensor and forwards data to the co-

ordination center. Some research was performed, in

order to ﬁnd an easily wearable device while walking

or running, and able to transmit its data using Blue-

tooth to a mobile device. In that sense, it was also

of vital importance the availability of the data proto-

col from the vendor. It was necessary to connect and

decode the stream of information coming from the de-

vice’s Bluetooth interface. The selected device was a

Nonin 4100 pulse oximeter.

A mobile phone with Android Operating System

was selected because of its openness and growing ac-

ceptance. The only requirements were to include a

Bluetooth module and GPS hardware in order to geo-

graphically locate the mobile user. For that purpose,

a mid-range phone, the HTC G1 was selected.

The objectives of this approach are presented in

section 2. In order to fulﬁll them, the system structure

of section 3 and the functional visualization in sec-

tion 4 are proposed. Experimental results are shown

in section 5. Finally, conclusions and future work are

presented in section 6.

2 OBJECTIVES

The System’s midterm objective is to proof the viabil-

ity of a mid-highrangecellular phone in order to mon-

itor some of the user’s physiological parameters. Our

target user will be someone with a cell phone capa-

374

Reveriego G., Tellez J., Urdiales C. and Sandoval F..

SENSECARE - Real-time Location-based Health Monitoring System.

DOI: 10.5220/0003160903740377

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2011), pages 374-377

ISBN: 978-989-8425-37-9

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 1: System global structure.

ble of interacting with the environment and be aware

of the context. This system aims to provide the cell

phone with a new dimension of awareness, so that it

is capable of interact and monitor the user’s physio-

logical parameters such as heart and respiratory rate,

oxygen saturation, etc.

The system will receive physiological data from

sensors placed on the user’s body. These sensors will

connect to the phone by Bluetooth technology. At

present, technology does not allow phones to receive

physiological information directly. Commercial off

the shell tools and hardware have been used in order

to cover the system’s requirements.

This system will use the available hardware in or-

der to allow the user and/or his/her carer real time vi-

sualization of physiological data while moving. The

carer can be described as the actor of the system who

needs to manage the user’s physiological information.

There might be some cases where the user will be

the athlete or sportsman/woman who wants to mon-

itor his/her physical progression. A more specialized

and attractive use is that given by a health centre, a

doctor o even a relative that, for medical reasons, re-

quires monitoring a person with speciﬁc needs. The

description of those needs goes far beyond the objec-

tives of this paper.

3 SYSTEM STRUCTURE

A near real-time monitoring and control system of a

moving user has been developed. Devices for cap-

turing personal physiological data are placed on the

user’s body. Those sensors connect wirelessly to the

mobile phone, that acts as a gateway to the coordi-

nation centre. For that purpose, a custom software

application has been developed on the HTC G1. The

coordination centre gathers the information captured

by the cell phone and offers it to any actor who may

require it. Vital signs coming from a moving user will

be displayed in real time through Google Earth.

Fig. 1 shows the system’s global architecture. The

following sub-systems can be distinguished:

• Mobile Platform. It gathers and manages ex-

ternal data coming from physiological sensors

placed on the user’s body by means of a Blue-

tooth technology and location information from

Geo-localization systems. The phone stores cap-

tured data in an internal database and periodically

sends the information to the coordination centre

by 3g.

• Coordination Centre. The coordination centre

can be described as the element in this system in

charge of processing and managing the data re-

ceived from the mobile platform. Its mission is

safe storage of user’s data. This centre also pro-

vides actors/carers with the required information.

• Carer. A carer can have access to the information

previously processed and managed by the coordi-

nation centre and have a graphic display of it. The

carer has access to this information, and can act

accordingly in case of emergency. On the frame-

work of this system, a graphic interface will be

provided so that the carer can see the temporal,

geographical and physiological evolution of the

user. For this purpose, Google Earth was used,

since it allows a 3 dimensional display of geo-

referenced information where monitored physio-

logical parameters can be shown.

SENSECARE - Real-time Location-based Health Monitoring System

375

4 FUNCTIONAL VISUALIZATION

Once physiological parameters are available in the co-

ordination center, these data can be used with dif-

ferent purposes. For example, it is posible to detect

abnormal patterns that may end in a serious affec-

tion and warn the user. However, our purpose is to

show the user/actor/carer the information in an easy-

to-understand way. It is advisable that the applica-

tion used to read the resulting report is free and well-

known by the user, e. g., Google Earth. The func-

tional visualization described here is based in one

designed previously for a vehicle telemetry system

(Reveriego et al., 2009). In order to show the user

(doctor, carer) someone’s activity, there are several

parameters that could be of interest such as temper-

ature, heart rate, breath rate, etc. It depends on the

application. In this experiment, the shown parameters

are the following:

1. Location;

2. Pulse rate;

3. Oxygen Saturation Level;

4. Risk Level.

The representation proposed here consists of a 3-

dimensional graph (see Fig. 2). It is composed by

a string of prisms. Those prisms are deﬁned using

the open standard KML and represent the information

showed in Table 1.

Figure 2: 3-dimensional representation in a middle-distance

running.

Table 1: Correspondence between Prism dimension and

representing magnitude.

Prism dimension Representing magnitude

Height Pulse rate

Width Oxygen Saturation Level

Color Risk level

The color of the prisms depends on the risk level.

The cooler the color the lower the risk level and the

warmer the color the higher the risk level. The level of

risk is calculated by simple thresholding of the mea-

sured heart rate (Fox et al., 1971). Each prism is a

Polygon

KML object.

5 EXPERIMENTAL RESULTS

Although the system is in an initial stage, positive out-

comes can be drawn from each of the three compo-

nents of the architecture described in section 3. The

design of the platform aims to be as transparent as

possible to the user, since it intends to be a gateway

between the user and the coordination centre. The ap-

plication must interact with several other systems at

the same time seamlessly:

• Visually, with the user in order to display the in-

formation sent by the sensors;

• Management of Bluetooth communications with

physiological sensor placed on the user;

• Management of user’s geo-localization;

• Sending of the information gathered to the coordi-

nation centre without the intervention of the user.

The application meets the above mentioned re-

quirements, although its main limitation lies on bat-

teries autonomy; the simultaneous use of the GPS, the

Bluetooth radio and the uploading of the data by 3g,

results in considerably high power consumption.

The operation of the coordination centre is ade-

quate and uses open source tools.

By using Google Earth API, the carer can visu-

alize on a real map a 3D representation of the user’s

physiological activity in real time. To make this pos-

sible, the coordination centre shall generate and up-

date the information to be shown on the client Google

Earth of the carer when starting. In short, when the

carer starts Google Earth, he/she will be able to vi-

sualize how the user’s physiological and geographical

information updates in real time. The carer will also

observe how some dynamic polygons of different col-

ors and shapes are drawn on the map.

Fig. 3 shows a screen capture of the carer’s inter-

face where a high speed running route made by the

monitored person is traced. At ﬁrst, the the heart rate

is slow as the graph is low in height. Probably, the

user was staying quiet or just walking. Afterwards,

the graph starts to increase rapidly till a higher level

and the color of the graph pass from blue, warmer and

warmer, to red. Thus, the activity of the user grows in

intensity. This is the graph generated by a user that

start running fast from a relaxed state.

BIODEVICES 2011 - International Conference on Biomedical Electronics and Devices

376

Figure 3: High speed running from relaxed state.

6 CONCLUSIONS AND FUTURE

WORKS

A system capable of monitoring physiological param-

eters in real time has been presented. It is, basically,

continuous remote monitoring system comprising:

• A sensor device placed on the user’s body that

measures physiological data (pulse, Sp02);

• A cellular phone that retrieves data from the sen-

sor device using a short range technology (blue-

tooth); it collects geographical information (GPS)

and it sends it through 3g technology to the coor-

dination centre;

• A coordination centre acting as a server that stores

physiological data coming from the cell phone.

The coordination centre also receivesthe demands

from the users that wish to visualize physiological

data.

It is concluded that the designed visualization sys-

tem is intuitive and efﬁcient. It is, however, tedious

to do a real time monitoring of the patient for a long

time. It is very common that neither a physician nor

a carer carries out a real time continuous monitor-

ing of the patient. In this case, it seems more useful

to implement a system that allows detecting anoma-

lies through data mining, for instance, and alerts doc-

tors, carers or even the patient if abnormalities are de-

tected. This system could be more useful in situations

like some kinds of rehabilitation or tests, like stress

test for heart disease, when human supervision is ad-

visable. For this purpose, the measurement of more

speciﬁc vital signs would be a requirement.

Another problem that has to be dealt with is the

battery life time of the cell phone. Phone battery life

is relatively short (2 hours), since a continuous mon-

itoring implies the use of hardware with high power

consumption (GPS, Bluetooth, 3g radio).

Future work will focus on three main lines. First,

behavior-based preventive alarms would be interest-

ing to warn users, carers and doctors. Second, it might

also be interesting to ﬁnd transparent, cheap, commer-

cial sensors that could be attached to the mobile phone

to improve risk detection and provideadditional infor-

mation. Finally, system software has to be optimized

to extend battery life.

ACKNOWLEDGEMENTS

This work has been supported by CENIT-AmIVital,

Ingenio 2010 project and by the Spanish Ministerio

de Educacion y Ciencia TEC2008-06374-C02-01.

REFERENCES

Fox, S. M., Naughton, J. P., and Haskell, W. (1971). Phys-

ical activity and the prevention of coronary heart dis-

ease. Annals of clinical research, 3(6):404–32.

Jones, V. (2006a). Mobihealth: mobile health services

based on body area networks. Centre for Telemat-

ics and Information Technology University of Twente,

Enschede.

Jones, V. (2006b). Mobihealth: mobile services for health

professionals. Centre for Telematics and Information

Technology University of Twente, Enschede.

Litan, R. E.(2008). Vital signs via broadband: Health moni-

toring transmits savings, enhances lives. Better Health

Care Together, pages 38–42.

Pandian, P. (2008). Wireless sensor network for wear-

able physiological monitoring. Journal of Networks,

3(5):21–29.

Reveriego, G., Urdiales, C., and Sandoval, F. (2009). Vehi-

cle risks assessment using a nokia n95. In Intelligent

Transportation Systems, 2009. ITSC ’09. 12th Inter-

national IEEE Conference on, pages 1–6.

Romero, E., Araujo, A., Malagon, P., Vallejo, J. C., Moya,

J. M., and Nieto-Taladriz, O. (2008). Patientsmonitor-

ing system based on a wireless sensor network adap-

tive platform. In pHealth. International Workshop

on Wearable Micro and Nanosystems for Personalised

Health, Valencia.

SENSECARE - Real-time Location-based Health Monitoring System

377