An Arduino based Health Monitoring System for Elderly People

Living at Home

Functions and Ontology

Alfio Costanzo and Carmelo Pino

Department of Electrical, Electronics and Computer Engineering, University of Catania,

Viale Andrea Doria 6, Catania, 95125, Italy

Keywords: Arduino based System, Health Monitoring Devices, Biosignal Acquisition, Analysis and Processing, User

Model Ontology.

Abstract: In recent decades, people, especially the older ones, try to live at home in autonomous way. For this purpose

it is useful to monitor their vital signs and the environment that surrounds them with the aim of activating

suitable environment regulation and when needed to send alarms to family members, medical, or hospitals,

according to the criticality of the subject. The paper aims at proposing a flexible and reliable monitoring

system based on Arduino shields. The main feature of the system is that it allows the doctor and family

members to monitor the patients at distance using their mobiles. A suitable communication with the first aid

center is foreseen for a fast rescue of the patients in case of critical situations. The general user model

ontology is used so that the personal data featuring the patients and the relevant context may be used by any

diagnostic and first aid software, thus envisaging an open and interoperable health monitoring system for

elderly people living at home.

1 INTRODUCTION

With the increasing number of elderly people willing

to live independently, it could be helpful to monitor

their behaviour and to check in real time their health

conditions. Also, it could be necessary to have

automated systems that adapt the environment

conditions to the people health status. Such a system

should inform the family or the doctor about the

possible critical health conditions, and should be

able to send alarms to the hospital including the user

health data and the best path to rescue the patient if

the conditions are very critical.

Several proposals are available in the literature,

e.g., (Huo H., 2009), but the patient data cannot be

downloaded, processed and visualized at distance by

using fixed and mobile devices, neither the home

environmental conditions are taken into account.

Therefore, aim of the paper is to show how a

monitoring system of the health status of elderly

people living at home may implemented by an

Arduino platform to carry out both a first diagnosis

of the health status and to adopt changes of the home

environment parameters, e.g., home temperature and

humidity, to improve life condition in the home.

Sect. 2 shows how the proposed E-Health

Monitoring System based on the Arduino platform

(Banzi M., 2008) may check the elderly people

health status. In the paper, we focus our attention on

the monitoring of parameters that cannot be made by

wearable sensors such as blood pressure and

respiratory rate, but it could be possible to extend

the monitoring to the other vital signs taken while

the patient is walking by using the system illustrated

in the companion paper (Pino C., 2014).

Sect. 3 illustrates how the E-Health System could

be expanded with an Environment Conditions

Monitoring System and how it is possible to change

the environmental home conditions according to the

health status and to alert family members, the doctor

or even the hospital just in case the environment

changes don’t produce the desired positive changes

of the monitored vital signs.

Sect.4 proposes to store all the mentioned

information according to the general user model

ontology (GUMO) (Heckmann D., 2005) so that the

personal data featuring the patients and the relevant

context may be used by any diagnostic and first aid

software independently on the computing system in

which the software resides, thus envisaging an open

336

Costanzo A. and Pino C..

An Arduino based Health Monitoring System for Elderly People Living at Home - Functions and Ontology.

DOI: 10.5220/0004762303360340

In Proceedings of the International Conference on Physiological Computing Systems (PhyCS-2014), pages 336-340

ISBN: 978-989-758-006-2

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

and interoperable health monitoring system for

elderly people living at home. Concluding remarks

illustrate open problems and future works.

2 E-HEALTH MONITORING

SYSTEM

The implementation of automated and intelligent

system that carries out the monitoring of the vital

parameters may enhance and support the quality of

life of the people, especially the elderly ones that, in

recent decades, are willing to live at home in

autonomous way. Our health monitoring system

architecture, drawn in fig.1-up, is able to measure

the blood pressure and the respiratory rate with

respect to the home environmental conditions. The

nucleus of the system consists of an Arduino Mega

shield powered by the e-health Arduino sensor

shield shown in in fig.1-down.

Figure 1: a) Arduino based monitoring system

Architecture tracks the user vital signs and sends this

information to their mobile. Also, it tracks environment

condition and, if is necessary, changes them according to

the health status of the user. b) E-Health Monitoring

System based on Arduino.

Other two portable monitoring devices, i.e., the

glucometer and the oxygen in the blood, may extend

the blood pressure and respiratory sensors taken into

account in the papers. Due to the reduced

dimensions and costs with respect to the ones

available in literature, e.g., (Islam M., 2012), all the

mentioned sensors are easily portable from a

location to another in the house, whereas other

wearable sensors could be added as illustrated in

(Pino C., 2014) to monitor the patient while moving

at home.

The main feature of the proposed system is that it

is able to send in real time the collected data to the

doctor, to the family members, or, if the patient data

are critical, to the nearest hospital using the simple

and reliable communication channels pointed out in

fig.1, i.e., through the GPRS/GSM and LAN shields

mounted at the top of the Arduino mega shield.

In both cases, the system is provided with an

internal memory where recent health data are stored.

In the GPRS case, the relevant data and alarms are

sent directly to the family members, to the doctor

and possibly to the hospital. In the LAN case, the

system sends data and alarms by means of a cellular

phone connected to the LAN.

Of course, the former solution is more reliable

and does not suffer of battery problems. The second

could not work if the cellular phone battery is low,

but it is more flexible since it may host a specific

diagnostic software that avoids both false positives

and negatives by executing further tests on the

patient conditions. Therefore, both GPRS and LAN

shields are recommended so that the system may

meet reliable and flexible requisites.

The functions to take the current values of the

systolic and diastolic pressures are embedded in the

systems and can be recalled as shown in fig.2.

Figure 2: Functions to retrieve systolic and diastolic

pressures.

The other functions to take into account to measure

the rate of airflow breathed in and out, as well as the

wave of the respiratory rate are shown in fig.3.

To visualize these values, the monitoring system

could be provided with a small display, but we have

implemented on the mobiles a software, developed

with Flash Builder (Corlan M., 2009), that is able to

display the above vital parameters, as shown in fig.4.

Let us note that the monitoring system is able to

send to a mobile the data according to the JSON

format within anonymous messages as envisaged in

(Anciaux N., 2013), but to generalize its

functionalities we have implemented three important

functions on the mobile: a) the conversion of the

received JSON strings to XML/RDF triples, b) the

storage of the triples into the mobile memory and c)

the triple visualization on the mobile display using a

int systolicPressure() {

return eHealth.getSystolicPressure();

}

int diastolicPressure() {

return eHealth.getDiastolicPressure();

}

AnArduinobasedHealthMonitoringSystemforElderlyPeopleLivingatHome-FunctionsandOntology

337

Figure 3: Functions to retrieve respiratory rate and airflow

wave.

Figure 4: Some vital parameters shown on the mobiles and

the related Json format.

software that is able to visualize the main values of

the user physiology according to a standard

XML/RDF ontology derived from the Json format

shown in fig.4-right, and inspired by (Faro A..,

2003), (Heckmann D., 2005).

This allows the doctors to visualize at distance

the current health status and the health history of the

patient using any type of mobiles provided that the

mobile is equipped with the proposed visualization

software.

3 ENVIRONMENT MONITORING

SYSTEM

Irregular respiratory rates or sudden changes in

respiratory rate are a broad indicator of a major

physiological instability; in many cases, the

incorrect respiratory rate is one of the earliest

indicators of this instability, e.g., (Ionescu et al.,

2009), (Cnockaert L., 2008).

Therefore, our system can provide an early

warning of hypoxemia and apnea. Also, monitoring

blood pressure at home is important for many

people, especially for the elderly patients who have

high blood pressure. However, both respiratory rate

and blood pressure should not always have the same

average value for the elderly people living at home.

Indeed, they may change depending on the

emotional status, on if the patient is walking, doing

exercises or sleeping, and on the home environment

conditions. Therefore, monitoring the main personal

conditions is also important, e.g., emotional status

could be identified recognizing modified facial

features or reduced visual attention abilities, e.g.,

(Faro A., 2010), (Faro A., 2006), even in noisy

contexts (Cannavò F., 2006), (Crisafi A., 2008).

Also, it may be detected by measuring the

perspiration activity, e.g., using galvanic sensors.

Many sensors are available to monitor the physical

activity and position, e.g., (Cooking-Hacks).

A complete analysis of the elderly overall status

is outside the scope of the paper and it is for further

study, whereas we are more interested in controlling

the home environment conditions since they may

influence greatly the blood pressure and the

respiration status, e.g. warm temperature can

increase blood pressure.

In many cases, a regulation by a simple

automatic system may be enough to restore the right

health status, e.g. the automated opening of a

window or the starting of an air conditioner. For this

reason the first check, carried out by our system, is

the one of comparing the vital parameters with the

surrounding environmental conditions.

Fig.5 shows a possible environment monitoring

system controlled by an Arduino mega shield that

takes information about light and temperature.

Figure 5: Environment Monitoring System.

This allows the system to control temperature

int respiratoryRate() {

return eHealth.getAirFlow();

}

void airFlowWave(int respiratoryRate) {

eHealth.airFlowWave(respiratoryrate);

}

PhyCS2014-InternationalConferenceonPhysiologicalComputingSystems

338

and light of the rooms and if the patient blood

pressure is high, to lower it by opening/obscuring a

window and starting an air conditioner. Only after

having observed that these actions are not

successful, the system will carry out further actions:

first, it will indicate to the patient, on his/her mobile,

the nearest hospital, then it will inform the family

members and the doctor about the patient health

status, and finally it will send the alarm to the near

hospital including information on the vital

parameters of the patient.

4 USER MODEL ONTOLOGY

Fig.6 shows a simplified scheme of the general user

model ontology (Heckmann D., 2005) that we

propose to use to collect all the relevant data

according to an open and interoperable data

organization so that people can be aided in either

normal or critical situation by cooperating intelligent

agents resident on different computing devices, e.g.,

the user or family doctor mobile, the computing

system located on the ambulance, or the server of

the hospital.

Figure 6: General User Model Ontology.

In such scheme, we point out how the data collected

by our health monitoring system are useful to fill the

sections related to physiology and environment. In

addition, we propose to substitute the section current

state by a section dealing with the current and

historical state consisting of records featured by

clock time and user status periodically added to the

previous ones.

This allows the doctors to know the

physiological data, e.g., the breath rate data with

respect to the clock time and to the status of the

people at this time. For example they may know the

heart data in the period in which the people is

suffering from anxiety. Of course, only a limited

amount of the data should be available on line, e.g.,

the data featuring the last hour relevant to plan the

best rescue in case of sudden critical user condition

could be stored on the user mobile, whereas the

previous data, more relevant for clinical purposes,

should be stored in the personal computer available

at the user home or at the hospital.

Fig.7 shows the XML/RDF format of the data

available on the computing system extracted by a

query issued by an intelligent software agent

resident either on the local monitoring system or at

distant ambulance or hospital servers to know the

current health status of the elderly people and of the

variable featuring the ambient.

Figure 7: Personal and environment data in XML format.

A straightforward conversion of the XML/RDF

format of the query response to OWL format is also

available in our implementation to represent the data

according to the web semantic standard format, i.e.,

according to the terminology and properties of well

established ontology approved by W3C such as

FOAF and SSN related to people and sensor

ontology. A deeper discussion on this subject is for

further study.

<blood>

</blood>

</breath>

</datetime>

</environment>

</health>

Systolic Pressure

Diastolic Pressure

Breath Rate

Hour

Minute

Second

Day

Month

Year

Health

Status

* Add Historical State

Temp.

Humidity

Pressure

Preferred Media, Job,

Holiday, Contents,...

Geo-coord.

Address

AnArduinobasedHealthMonitoringSystemforElderlyPeopleLivingatHome-FunctionsandOntology

339

5 CONCLUDING REMARKS

The paper demonstrates the feasibility of reliable

and flexible health monitoring systems aiming at

controlling relevant health parameters by means of

portable sensors of small dimensions, e.g., the ones

needed to measure the blood pressure, the

respiration rate and the glucose levels.

The reduced dimensions and the low cost of the

selected Arduino shields and their communication

features makes possible the monitoring at distance of

the elderly people living at home using smart

phones.

The RDF format chosen to store, on the memory

of the Arduino GPRS and LAN shields, the data

coming from the sensors allows the servers of

authorized people to download and interpret the

patient data with simple RDF based procedure.

The use of an agreed ontology such as GUMO is

envisaged to open the data stored on the Arduino

shield to the devices of remote authorized people. In

particular, this will make possible to visualize the

patient data on any remote mobile (of the family

doctor or the user relatives) where they are displayed

using simple procedure or integrated to other

software able to carry out some deep diagnosis

aimed at activating specific interventions.

We plan to use this system into the WiCity

Project whose general aim is the one of supporting

mobility, logistics and user health assistance in a

smart city, e.g., (Costanzo A., 2012), (Faro A.,

2011), (Faro A., 2008). A more powerful decision

support system to improve people health status is

planned according to the lines indicated in (V.L.

Sauter, 2011), (Costanzo A., 2013).

REFERENCES

Huo H., et al.: An Elderly Health Care System Using

Wireless Sensor Networks at Home, 2009 Third

International Conference on Sensor Technologies and

Applications, IEEE Computer Society, 2009.

Banzi M., Getting Started with Arduino, Make 2008.

Pino C., Costanzo A., Arduino based System for

Monitoring the ECG Status and Improving Home

Environment Conditions Int. Conf. on Physiological

Computing Systems, PhyCS, Lisbon, 2014.

Heckmann D., et al., GUMO, the General User Model

Ontology, LNCS, N.3538, Springer, 2005.

Islam M., et al., Development of a Noninvasive

Continuous Blood Pressure Measurement and

Monitoring System, IEEE/OSA/IAPR Int. Conf. on

Informatics, Electronics & Vision, 2012.

Corlan M., Flash Platform Tooling: Flash Builder, Adobe, 2009.

Anciaux N., et al. Personal Data Management with Secure

Hardware - The Advantage of Keeping your Data at

Hand, IEEE Proc on Mobile Data Management

(MDM), Milano, 2013.

Faro A., Giordano D., Musarra A., Ontology Based

Mobility Information Systems. Proc. of IEEE Systems,

Men and Cybernetics SMC03, vol.3, 4288-4293, 2003.

Ionescu et al., Mechanical Properties of the Respiratory

System Derived From Morphologic Insight, IEEE

Transactions on Biomedical Engineering, Vol. 56(4),

2009.

Cnockaert L., et al., A Method for the Analysis of

Respiratory Sinus Arrhythmia Using Continuous

Wavelet Transforms, IEEE Transactions on

Biomedical Engineering, Vol. 55(5), 2008.

Faro A., Giordano D., Pino C., Spampinato C., Visual

attention for implicit relevance feedback in a content

based image retrieval, Proc. ACM Symposium on Eye-

Tracking Research and Applications, ETRA 2010.

Faro A., Giordano D., Spampinato C., An automated tool

for face recognition using visual attention and active

shape models analysis, Proc. 28th Annual

International Conference of the IEEE Engineering in

Medicine and Biology Society, EMBS'06, 2006.

Cannavò F., Nunnari G., Giordano D., Spampinato C.,

Variational Method for Image Denoising by

Distributed Genetic Algorithms on GRID

Environments. Workshops on Enabling Technologies:

Infrastructure for Collaborative Enterprises, WETICE

'06. 227-234, IEEE, 2006.

Crisafi A., Giordano D., Spampinato C., GRIPLAB 1.0:

Grid Image Processing Laboratory for Distributed

Machine Vision Applications. Proc. Int. Workshop on

Enabling Technologies: Infrastructure for

Collaborative Enterprises, WETICE '08. 188-191,

IEEE, 2008.

Cooking-Hacks, Patient Position Sensor for e-Health

Platform, http://www.cooking-hacks.com.

Costanzo A., Faro A., Giordano D., Venticinque M., Wi-

City: A federated architecture of metropolitan

databases to support mobile users in real time, IEEE

Proc. Int. Conf. on Computer & Information Science

(ICCIS), Vol.2, 586-591, 2012.

Faro A., Giordano D., Spampinato C., Integrating

Location Tracking, Traffic Monitoring and Semantics

in a Layered ITS Architecture. Intelligent Transport

Systems, vol.5

(3), 197-206, 2011.

Faro, A., Giordano, D., Spampinato C., Evaluation of the

traffic parameters in a metropolitan area by fusing

visual perceptions and CNN processing of webcam

images, IEEE Trans. on Neural Networks, Vol.19(6),

pp.1108-1129, 2008.

V. L. Sauter, DSSs for business intelligence, Wiley, 2011.

Costanzo A., Faro A., Giordano D., Wi-City: living,

deciding and planning using mobiles in Intelligent

Cities, 3rd International Conference on Pervasive and

Embedded Computing and Communication Systems,

PECCS, Barcelona, INSTICC, 2013.

PhyCS2014-InternationalConferenceonPhysiologicalComputingSystems

340