MOBILE HEALTH MONITORING

PLATFORM FOR AAL APPLICATIONS

J. Festa, C. Silva and P. M. Mendes

Centro Algoritmi, University of Minho, Campus de Azurém, 4800-058, Guimarães, Portugal

Keywords: Mobile Health, Ambient Assisted Living, Vital Signs Monitoring.

Abstract: Mobile devices able to monitor diverse health condition parameters are becoming widely available. Also,

mobile devices with wireless communications with significant computing capability (e.g., tablets,

smartphones) are becoming available every time everywhere. Placing such devices to operate together will

allow to deploy the so called ambient assisted living technologies, which allows for individual health

condition monitoring almost everywhere and any time required. This paper presents a solution able to

monitor several physiological parameters using the mobile platforms running android. The implemented

solution permits to integrate transparently services from remotely distributed devices.

1 INTRODUCTION

The use of information and communication

technologies is nowadays widely accepted as a way

to offer new solutions for healthcare. Those

solutions will improve the quality of life of patients,

reducing costs at the same time. However, before

that happens, new solutions, including both

hardware and software, must be available in order to

acquire and store the required signals, to process and

extract information from those signals, and to detect

a set of features required to fire alarms and/or

electronic assistance.

Despite the availability of several platforms for

signal acquisition (Pantelopoulos and Bourbakis,

2010), a few technological issues must be solved

before they can be used. The platforms must offer

quality of service, be wearable, and operate for a

comfortable period of time. The systems should

allow to remotely monitor one or a set of subjects

and to monitor their health and activity. Based on

that, and on a set of information associated to that

subject, it is possible to offer different levels of

services. One will be self-remainders that are preset

by the subject; the other will be reminders by some

familiar or caregiver; and finally, a health service

based on clinical data may also be offered, if good

enough health engines are available to extract the

“health condition”.

Despite highly desirable, the success of such

technologies depends on the users’ (patients,

caregivers, medical doctors) technology

acceptability. One way to increase the acceptability

is to design the solutions based on devices already

widely familiar to users.

One solution already proposed is to use android

based platforms. Such solutions were already used to

record physiological signals and make them

remotely available (Altini et al, 2010; Colunas et al,

2011). However, such solutions were not designed to

support a significant number of signals, patients, and

caregivers supplying and accessing to physiological

signals.

We present the architecture of a complete system

for a distributed monitoring system for ambient

assistant living. We also describe the components of

one of the subsystems, an application for a

smartphone and the supporting hardware for vital-

sign acquisition.

The paper is structure as follows: in the second

section, we describe the complete architecture for

our distributed system and its main subsystems. In

the third section, we describe the hardware of our

acquisition system. The fourth section, describes the

Android application that implements the components

of the mobile subsystem. Finally, we conclude.

2 e-UMIHEALTH SYSTEM

In Figure 1, we represent the application’s scenario

targeted by our system. In this scenario, we identify

341

Festa J., Silva C. and Mendes P..

MOBILE HEALTH MONITORING PLATFORM FOR AAL APPLICATIONS.

DOI: 10.5220/0003784903410344

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2012), pages 341-344

ISBN: 978-989-8425-91-1

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

three key elements of our system: an e-UMIHealth

domain controller (care centre), an e-HealthDroid

device and a bio-signal device.

An e-UMIHealth domain controller is a caregiver

centre. It is the entity in the network that manages

the e-HealthDroid devices. A key aspect of the

management is the definition of the users’ profiles.

These define categories of users that have different

access to the data, such as medical doctors, nurses or

other personnel. The e-HealthDroid device is a

smartphone based on the Android OS that

implements our software stack. Through its

application-programing interface (API), it provides a

bridge to its associated bio-signal device. In this

sense it works as server for the sensors’ data

acquired by the bio-signal device. The bio-signal

device is the sensing element that acquires the

several biological signals. A bio-signal device is also

called a secondary entity, since its sensors’ data is

only accessible through an e-HealthDroid device. An

e-HealthDroid device may have no bio-signal device

associated to it, but be used to inquire data on

another e-HealthDroid device (server). The e-

UMIHealth domain controller and the e-HealthDroid

device are also called primary entities, since they

interact directly among them.

Figure 1: Business use case.

2.1 Architecture

In Figure 2, we present e-HealthDroid architecture.

The e-HealthDroid subsystem has five components:

service manager, event manager, subscription table,

profile table and user table.

In the following subsections, we describe the

main components and their interaction.

2.2 Service Manager

The service manager manages all incoming service

requests. These requests are session based; therefore,

the manager requires that the requesting entities

must authenticate themselves before issuing any

further request besides the authentication service.

After opening a session, several services may be

issued till the session is closed. Upon receiving an

authentication service request, the service manager

contacts its e-UMIHealth domain controller to

validate the credentials of the user and its profile.

This information is recorded in the user table and it

is active while the session is open.

Figure 2: e-HealthDroid architecture.

Table 1: Description of services.

Service Description

Authentication

The authentication service is used by a

primary entity to authenticate an user in

order to request further services. All

services are subject to authentication

and are subject to the level of access

defined for the user’s profile.

Subscription

This service permits that a primary

entity may subscribe to one or the

several sensors’ data. This subscription

can be issued for a single value

(pressure at the moment) or a stream or

chunk of data (ECG signal).

Discovery

Any primary entity may get a list of all

kinds of sensors present in the bio-

signal device.

Localization

Using the localization service, we can

find the position of the e-HealthDroid

device. This service uses the

capabilities provided by the Android

OS.

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

342

When a primary entity requests a subscription

service, the service manager updates its subscription

table, taking into account the user’s profiler. This

information is used by the event manager. In Table

1, we list all the services provided by the service

manager.

2.3 Event Manager

The event manager is responsible for routing the

sensors’ data (events) according to the information

contained in the subscription table. The routing is

based on the publisher/subscriber model

(Buschmann, 1996). The data may be requested as

different types of events: a single sample event, a

chunk of samples event or a stream of samples

event. Upon finishing serving a single sample or a

chunk of samples, the event manager clears the

respective entry in the subscription table. An entry in

the subscription table may also be cleared by the

service manager if the associated session is closed.

3 ACQUISITION SYSTEM

In addition to the developed software, and for test

purposes, it is also required to implement an

acquisition system able to be worn by users. One

solution is to develop regular clothes with the

acquisition devices directly embedded. The other is

to develop small enough devices to be hidden in

regular clothes, with the ability the record the

required signals. Figure 3 shows the available device

used for signals recording (bio-signal device).

Figure 3: Physiological acquisition device.

This in-house device allows recording one lead

ECG, oximetry, and temperature. This device has

then possibility to be directly connected to an e-

HealthDroid device (android platform) through USB

or using a wireless link. Either case, the recorded

signals may become available through the e-

HealthDroid device.

4 THE APPLICATION

Figure 4 presents the various screens for the Android

application that implements the API of the e-

HealthDroid sub-system. This application enables

the user to connect to his bio-signal device or other

servers (e-HealthDroid devices), choose events from

the ones permitted by each server and see the

received data in the form of a dynamic graph or a

numerical value. The e-HealthDroid device can itself

be a publisher for the user’s events (associated bio-

signals device) and so permissions should be defined

in the profiles.

The principal screen (Figure 4a) is the one where

the user fills the user and password’s login and

writes the name of a server to which is intended to

connect. The login is necessary so that the server can

obtain the client’s category (by contacting the Care

Center Database) and tell the client what events can

be then subscribed. The Figure 4b) shows the

Subscribing screen, where the user selects one of the

servers to which is already connected and is

presented with the events that has permission to

subscribe. In the exampled case, the user selected

the Festa-HP device, for which has permission to

subscribe the ECG, the Heart Rate, the SPO2, the

Temperature and the Server’s Profile. In this case,

the user is going to subscribe the ECG event. After

this subscription the user is conducted to the

graphical screen (Figure 4c), the most important

screen, for the visualization of the ECG signal in

real time, sent by the server’s device.

Every screen has at least two options: the

configuration and the application exit. The

configuration option leads to the permissions area

(Figure 4d and Figure 4e) where the user can define

the permissions for each category of clients

(profiles) who are allowed to subscribe each event.

This can be done by first selecting the event and

then choosing its desired corresponding permissions

from the categories defined as Everyone (in case the

event could be subscribed for every person

connected), Doctor, Nurse, Other Medical Staff

(such as an auxiliary or a medical student) and Care

Center.

Other options are the Subscribing, which lead to

the subscribing screen to add more events to the

subscribed ones, and the unsubscribing screen,

where the user is conducted to a similar screen to the

previous one, but where the events presented are the

ones already subscribed and where the user selects

the ones to stop subscription. The last option is Start

Menu, where the user goes back to the principal

screen, with a new connection opportunity.

MOBILE HEALTH MONITORING PLATFORM FOR AAL APPLICATIONS

343

(a) (b)

(c)

(d) (e)

Figure 4: UMi-HealthDroid application screens: (a) Start

screen – connection to servers (with Android Menu button

pressed), (b) Subscribing screen, (c) Graphical screen in

landscape, (d) Permissions screen (to selected event), (e)

Permissions screen (to define permissions for the selected

event) for each profile.

5 CONCLUSIONS

In this paper, we presented a distributed system for

vital-sign acquisition. It permits that health

professionals may monitor remotely and

transparently critical information, as well to locate

the patient using the capabilities provided by current

smartphones based on Android OS. Another

important aspect is the use of permission profiles

that allows customising the access through a central

authentication that ensures privacy of information.

REFERENCES

Altini, M., Penders, J., Roebbers, H., 2010. An Android-

Based Body Area Network Gateway for Mobile

Health Applications. Wireless Health ’10. San Diego,

pp 188-189.

Buschmann, F., Meunir, R., Rohnert, H., Sommerlad, P.,

Stal, M., 1996. Pattern-Oriented Software

Architecture: A System of Patterns. John Wiley.

Colunas, M., Fernandes, J., Oliveira, J., Cunha, J., 2011.

Droid Jacket: Using an Android based smartphone for

Team Monitoring. 7th International Wireless

Communications and Mobile Computing Conference –

IWCMC 2011. Istambul.

Pantelopoulos, A., Bourbakis, N., 2010. A survey on

wearable sensor-based systems for health monitoring

and prognosis. IEEE Transactions on Systems, Man,

and Cybernetics, vol. 40, pp 1-12.

BIODEVICES 2012 - International Conference on Biomedical Electronics and Devices

344