Using Smartphone Bases Biodevices for Analyzing Physiological,

Psychological and Behavioral User’s Habits

Federico Guede Fernández

, Marc Pous Solà

, Miguel Ángel García González

Lluís Capdevila Ortís

, Juan Ramos Castro

and Mireya Fernández Chimeno

Universitat Politècnica de Catalunya, Campus Nord C4 Jordi Girona 1-3, Barcelona, 08034, Spain

Universitat Autònoma de Barcelona, Campus de la UAB Edifici B Bellaterra, Cerdanyola del Vallès, 08193, Spain

Keywords: Android, Real-time, Biomedical Monitoring, Biosensor.

Abstract: As a consequence of increasing life expectancy, the promotion of lifestyles that allow aging wellbeing

guarantees has acquired great importance in the developed countries. However, the adherence to healthy

behaviors in young and adult people remains as a big problem in the community health field. The

development of markers of adherence to healthy lifestyles and the evaluation its effectiveness is a goal of

many research groups. This paper presents a system for analyzing physiological, psychological and

behavioural user’s habits using a smartphone and externals biodevices. We use an Android smartphone with

an internal tri-axial accelerometer and GPS to monitor physical activity. The smartphone is connected via

Bluetooth to a respiratory sensor for breath monitoring. In addition, Android application contains

psychological questionnaires to analyze user’s mood state and at the same, social interaction is analyzed

tracking phone usage and user’s social network. Finally, the collected information is sent to a remote server

for a long-term processing.

1 INTRODUCTION

As a consequence of increasing life expectancy, the

promotion of lifestyles that allow aging wellbeing

guarantees has acquired great importance in the

developed countries. However, the adherence to

healthy behaviors in young and adult people remains

as a big problem in the community health field. The

development of markers of adherence to healthy

lifestyles and the evaluation its effectiveness is a

goal of many research groups.

Healthy lifestyles includes among others, regular

exercising and daily physical activity, eating and

sleeping habits, and stress control. In order to

measure the adherence to these healthy lifestyles, is

necessary to have non-intrusive measurement

devices, almost transparent for the users, and

measurement protocols suitable for every particular

application.

The application fields of healthy lifestyle

adherence measurements range from monitoring old

people habits to the sportsmen and sportswomen

quality of training measurement. They also include

driver somnolence control, quality of life of workers

in shift, or simply individual self-help applications.

Many handheld mobile devices have high speed

data transmission capabilities (e.g., 3G, WiFi) and

embedded microprocessors with the capability to

communicate, using wireless connection (e.g.,

Bluetooth, NFC), with external sensors. Moreover

mobile devices have internal sensors (e.g.,

accelerometer, gyroscope), supplied by powerful

processors and large storage capacities. So,

smartphones are strong candidates to be used as

biodevices for monitoring parameters related with

healthy lifestyle behaviors. From the monitoring

point of view, the main advantage of mobile devices,

compared to laptops or desktop computers, is that

these devices are always handled by the user

collecting data during the whole day.

The smartphone market has grown, from being a

niche market product segment, becoming a truly

mass-market. In 2011, total annual global shipments

of smartphones exceeded those of client PCs

(including pads) for the first time (Canalys, 2012).

In fact, in Europe the smartphone penetration have

grown just 47.6% (comScore, 2012). Regarding

operative systems in smartphones, Gartner research

shows Android’s handset annual sales reached 82,1

million units in first quarter of 2012, in comparison

243

Guede Fernández F., Solà M., García González M., Capdevila Ortís L., Ramos Castro J. and Fernández Chimeno M..

Using Smartphone Bases Biodevices for Analyzing Physiological, Psychological and Behavioral User’s Habits.

DOI: 10.5220/0004206802430248

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2013), pages 243-248

ISBN: 978-989-8565-34-1

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

with last year sales represents a growth of 123%.

This means that Android accounted more than half

of all smartphone sales (56.1%) (Gartner, 2012).

In the area of medicine preventive there some

studies about the relation among obesity, nutrition

and physical activity (Hensrud, 2010). Another

study concludes that sleep loss is associated with

having an unhealthy lifestyle.(Ohida et al., 2001).

In several mobile health-care applications,

mobile devices are used as gateway for receiving

bio-information from medical portable devices and

mobile sensors. There, mobile device is responsible

of collecting data from sensors, pre-processing the

raw signals and transferring the gathered data to

remote server (Yang & Gerla, 2011).

For daily mood assessment, there is an android

application that combines accelerometer, sound

sensor, light sensor, GPS, and SMS information

(Yuanchao et al., 2012). In some applications,

mobile device also processes the gathering data and

only when mobile detects an emergency, it sends an

alert to a server or to social contacts (Sposaro &

Tyson, 2009).

Internal sensors of smartphone are used to

monitor heart rate variability using internal

accelerometers and recording the seismocardiogram

signal (Ramos-Castro et al., 2012).

This work exploits the increasing capabilities of

smartphones and its success between the people to

promote healthy life habits in people. The purpose of

this work is to develop an Android application to

analyze physiological, psychological and behavioral

personal habits. First we have to collect data to get

information about user’s life habits. Then

application shows some data to user and some

information will be sent to a server in order to a

third-party can process this information and monitor

the level of healthy habits.

2 MATERIALS AND METHODS

2.1 Android OS

Android is an open-source software stack created for

mobile phones and other devices produced by

Google. The Android SDK provides the API

libraries and developer tools required to build, test,

and debug applications. Android applications are

written in Java programming language and they run

on the Dalvik virtual machine optimized for mobile

devices. Android gives an open marketplace for

distributing it to Android users called Google Play.

Application components are the essential building

blocks of an Android application.

There are four different types of application

components: Activities, Services, Broadcast

receivers and Content providers. Each type serves a

distinct purpose and has a distinct lifecycle that

defines how the component is created and destroyed.

Our application uses activities, for user-interaction,

and a broadcast receiver that responds to system-

wide broadcast announcements, like Bluetooth

device have been found. Activities start on the main

application thread, therefore blocking I/O operations

or time-consuming processing have to be executed

in our own threads to avoid blocking the activity.

The layout is the architecture to design the user

interface (UI) in an activity. Layouts may be

declared using XML files or using Java code in

runtime. The main advantage of using XML files to

design the UI is that enables to separate the

presentation tier of the Activity logic tier that

controls its behavior (Google Inc., 2012a). This

improves the maintenance for later works.

Therefore, our application uses XML files.

We have selected Android OS to develop our

application for the following reasons: Android is an

open-source software and development tools are free

of charge; it also allows to interact easily with

internal sensors such GPS to track physical activity,

accelerometer to activity monitoring; it provides

simplified and well-documented Bluetooth, WiFi

APIs to communicate with external sensors.

Moreover there are several APIs to develop quickly

powerful applications.

2.2 Internal Sensor

Nowadays, the smartphones have integrated many

internal sensors, such as accelerometer, magnetic

field sensor, light sensor, GPS and proximity sensor.

The main advantage of these sensors is that they

don’t require an external communication channel,

such as Bluetooth of WiFi improves the battery

consumption and the facility to access the data

provided by the sensor.

For example, the accelerometer has been

employed in several applications to detect human

movement detection, to characterize physical

activity (Shin et al., 2010) recognizing if the user is

walking, going upstairs or running (Brezmes et al.,

2009). In our application we have developed filters

to reduce noise; Butterworth second order high-pass

filter is implemented. The accelerometer gives three

axis forces applied on the smartphone, in the

implemented application the contribution of earth

gravity have been removed.

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

244

Another sensor available in most of the mobile

phone is the light sensor. This can be used to detect

user light exposure. Nevertheless this sensor is not

useful to measure the level of sun exposure because

if the mobile phone is in the pocket or bag, do not

provide information of user exposition. For this

reason this sensor isn’t useful for this propose.

So, in function of the application, several

external sensors must be connected to the device, but

this is not a problem with the level of connectivity of

these devices

2.3 Breathing Monitoring

Breathing is a good indicator of somnolence

(Rodríguez-Ibañez et al., 2011). In order to record

the respiration signal we use an external monitoring

system (Bioplux Respiband sensor). It consists in a

respiratory sensor based on an inductive band to

measure thoracic effort. This sensor has the

embedded electronic to calculate breath signal.

Moreover, it has a Bluetooth communication module

that makes possible to connect it to smartphone.

Then, mobile application may monitor the breath

signal in real time. Mobile device parses data

received and displays breath signal to user.

Bluetooth communication uses Serial Port Profile

(SPP) that defines how to set up virtual serial ports

and connect two Bluetooth enabled devices. Once

Respiband has been connected to smartphone, it

starts to send data continuously to a frequency

configured between 36 Hz and 1000 Hz. In our

application the sampled frequency selected is 40 Hz.

This frequency is enough to characterize properly

the breath rate signal and reduce power consumption

due to Bluetooth transmission.

Android application has an activity to manage

Bluetooth connection. This activity lists paired

devices and allows to scan new devices, to pair and

to connect. Once paired an own thread is started to

establish RFCOMM channel and to maintain the

communication. This thread is continuously reading

new incoming data in real-time. When an incoming

packet is detected, the application parses it to extract

one byte of CRC and three bytes of value data.

Firstly CRC is checked to verify data integrity and

secondly, the application sends measured breath

value to UI to plot it. The data is visualized using an

open source charting library called AChartEngine

(4ViewSoft, 2012).

If the Bluetooth connection is

lost, the application detects it and tries to reconnect

continuously since user exit

2.4 User Location Monitoring

Android application uses location provider to track

exercise, using speed data, distance traveled and the

altimeter.

Android OS provide different methods to

determine the device’s current location, they are

called location providers, and each has its different

characteristics. Each location provider offers

different capabilities, including differences in power

consumption, accuracy, and the ability to determine

altitude or speed or heading information. The three

different locations providers are GPS_Provider,

Network_Provider, Passive_Provider. Therefore

each one is suitable to work over different

conditions. Using the Android Location Manager

API we can find available location providers, obtain

device’s current location, track movement, set

proximity alerts for detecting movement into and out

of a specified area.

GPS_provider determines location using

satellites. Depending on conditions, this provider

may take a while to return a location fix, this time is

called Time to First Fix (TTFF.) This time can be

cut down substantially by employing Assisted GPS

(A-GPS). A-GPS uses the mobile network to

transmit the precise GPS satellite orbit and clock

information to a mobile device. Then the device will

know the approximate location of the GPS satellites

in its line of sight. Reducing the number of

frequency bins which must be searched to acquire

the signal reduces the TTFF (Jarvinen et al., 2002).

The main limitation of GPS_provider is that it

only can work in outdoor environment, because in

indoors GPS signals don't reach. Moreover this

provider consumes more battery than the others.

Furthermore, GPS_provider provides the most

accurate location data and it is unique that has ability

to determine altitude information. Therefore, we

need use GPS_provider to track people exercise

properly.

Our Android application has a dedicated activity

that employs GPS_provider to obtain the user’s

location. Using information of location obtained, in

real-time, our application calculates main parameters

concerning a healthy behavior such as total traveled

distance, average speed and time spent. Apart from

displaying the parameters explained before the

application shows the route followed by the user

using maps. There are several libraries that provide

maps, we used Google Maps API (Google Inc.,

2012b) and an open alternative called OSMDroid

(Gramlich, 2012) to display OpenStreetMap (OSM).

The usage of both APIs is quite similar for

UsingSmartphoneBasesBiodevicesforAnalyzingPhysiological,PsychologicalandBehavioralUser'sHabits

245

programmer. Android application uses overlays to

draw followed route with points and lines. The main

disadvantage of Google Maps that it has a prize

policy of use otherwise OSM is totally free of use.

We have considered that more times while user

is doing exercise he is looking away from the screen.

Then for a better human-interaction Android

provides text-to-speech, application speaks to user to

notify each certain time period distance traveled and

time spent. Text to speech API provided for Android

it is used to implement this.

2.5 Social Interaction Tracker

Social interaction is important to human health. In

fact, individuals with adequate social relationships

have a 50% greater likelihood of survival compared

to those with poor or insufficient social

relationships. The influence of quality and quantity

of individual’s social relationships on the risk of

death are comparable with well-established risk

factors for mortality such as smoking and alcohol

consumption and they exceed other many risk

factors such as physical inactivity and obesity.

(Holt-Lunstad et al., 2010).

On the other hand, one of the most important

concerns associated with mobile phone use is that it

may become uncontrolled or excessive, which has

an impact upon daily living. Sleep disturbance is

among the most common negative outcomes

resulting from overuse of the mobile phone (Thomée

et al., 2011). Furthermore, smartphones allows

people to engage in a wide range of online

activities—such as Internet surfing, managing email,

playing video games or involvement in social

networks (e.g., Facebook, Twitter). Therefore,

mobile phone addiction should be conceptualized

within a broader spectrum of “cyber addictions” that

covers a wide range of behaviors relying on online

activities and/or activities involving communication

among individuals through technological devices.

Social network addictions is one of the most

frequent “cyber additions” (Billieux, 2012).

Android application can collect some

information about social interaction. On the one

hand, in order to track incoming and outgoing phone

calls and SMS, the application registers

BroadcastReceiver to track the phone usage. This

information is useful to evaluate social relationship.

Nowadays new social networks based on Internet are

very used to social interaction; Facebook and

Twitter are currently the most popular ones. Our

application uses Facebook (Facebook, 2012) and

Twitter API (Yamamoto, 2011) for Android parsing

JSON data received to collect information about

user’s social network like number of friends,

frequency update status, messages.

2.6 Psychological State Evaluation

Professionals and researchers are working in the

relationships existing between physical activity and

mood state. The most popular instrument for mood

state evaluation is POMS (McNair, 1984).

Questionnaire is a subjective method to evaluate

mood sate. Android application implements a test of

POMS with fifteen questions to evaluate user´s state

of mood obtaining the components of stress,

depression, hostility, vigor, and fatigue (Andrade et

al., 2010). The user has to indicate how the user

feels in 0-10 scale where 0 means not all and 10

very. There are three questions for mentioned

component to avoid random answers. In the

beginning and at the end of respiratory acquisition

the application launches the questionnaire to be

filled by the user. The results of the test answered by

the user provide useful information to study the

relationship between psychological, physiological

and behavioral habits.

3 RESULTS

Android application will forward physiological,

psychological and behavioral data obtained to a

remote server for a long term processing via

Android TCP socket when application finishes. It

uses Wi-Fi connection to reduce connection cost and

to increase rate transmission.

The breath signal acquired using Respiband

external sensor is monitored in real-time by the

application developed. In Figure 1, a screen captured

shows a XY graph displaying in y-axis the value of

amplitude and x-axis the samples. This graph is

refreshed when a packet of data is received. Usually

there is a packet for each sample of Respiband; its

sample rate is 40 Hz. The designed layout included a

graph to plot respiratory signal on time domain in

real-time. To plot the data the open source charting

library called AChartEngine is used.

For user’s location monitoring, application

shows the route tracked displaying it over a map. In

our application we have used two technologies that

provide maps: GoogleMaps API (Figure 2a) and

OpenStreetMap API (Figure 2b). In addition, the

application calculates and displays information

about distance traveled, average speed and time

spent. In order to improve the accuracy of this

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

246

information, user can pause and resume the tracking,

this is essential to avoid wrong results.

Figure 1: Mobile application screenshot: graphical breath

signal represented in real-time.

(a) (b)

Figure 2: Screenshots: route tracked using Google Maps

(a). Route tracked using OpenStreetMap (b).

The application has provided us useful types of

data to analyze social activity. In the application the

number and time when an SMS is sent or received

has been monitored. It also has allowed tracking the

duration, the number and the time of outgoing and

incoming phone calls. Concerning the social

networks, the application has been capable to

analyze the activity on Facebook and Twitter. The

application has obtained the number of friends that

is associated with the contact network size. In

addition to measure the social activity, the wall

panel, messages and notifications have been

checked. On Twitter social network published tweets

by the user, mentions, direct messages, the number

of followers and “following” have been quantified.

To evaluate the mood state in the application a

test with fifteen questions is launched in the

beginning and at the end (Figure 3a). The five

components of the mood state mentioned in section

2.6 have calculated each time when the user

completes the questionnaire (Figure 3b).

(a) (b)

Figure 3: Screenshots: a question of the test to evaluate the

mood state (a). Graph that represents the five components

resulted of a completed test (b).

The results obtained are stored in a database

proving us information of the psychological state

when the user is monitored.

4 CONCLUSIONS

A smartphone based on biodevice for healthy

lifestyle adherence evaluation has been developed.

The Android application developed collects the

information about different variables (activity,

respiration, social activity, etc.) and processes them

to measure the adherence to healthy lifestyles. The

data are sent to a remote server for further

processing to get results to improve user lifestyle.

ACKNOWLEDGEMENTS

This work has been partially funded by the Spanish

MINISTERIO DE CIENCIA E INNOVACIÓN.

Proyecto IPT-2011-0833-900000. Healthy Life style

and Drowsiness Prevention-HEALING DROP.

REFERENCES

4ViewSoft. (2012). AChartEngine: Charting Library for

UsingSmartphoneBasesBiodevicesforAnalyzingPhysiological,PsychologicalandBehavioralUser'sHabits

247

Android (Version 1.0.0) [Software]. Available from

http://code.google.com/p/achartengine/downloads/list

Andrade, E., Arce, C., Torrado, J., Garrido, J., De

Francisco, C., and Arce, I., (2010). Factor structure

and invariance of the POMS mood state questionnaire

in spanish. Spanish Journal of Psychology, 13(1), 444-

452.

Billieux, J., (2012). Problematic use of the mobile phone:

A literature review and a pathways model. Current

Psychiatry Reviews, 8(4), 299-307. doi:

10.2174/157340012803520522.

Brezmes, T., Gorricho, J. L. and Cotrina J., (2009).

Activity recognition from accelerometer data on a

mobile phone. In Proceedings of the IWANN ’09, 796–

799.

Canalys, (2012). Smart phones overtake client PCs in

2011. Retrieved February 3, 2012, from

http://www.canalys.com/newsroom/smart-phones-

overtake-client-pcs-2011.

comScore, (2012). Number of european smartphone users

accessing news surges 74 percent over past year.

Retrieved March 22, 2012, from

http://www.comscore.com/Insights/Press_Releases/20

12/3/Number_of_European_Smartphone_Users_Acces

sing_News_Surges_74_Percent_Over_Past_Year.

Facebook, (2012). Android Tutorial. Retreived May 18,

2012, from https://developers.facebook.com/

docs/mobile/android/build/

Gartner. (2012). Gartner says worldwide sales of mobile

phones declined 2 percent in first quarter of 2012;

previous year-over-year decline occurred in second

quarter of 2009. Retrieved May 16, 2012, from

http://www.gartner.com/it/page.jsp?id=2017015.

Google Inc. (2012a). Android developer guide: Layouts.

Retrieved May 11, 2012, from

http://developer.android.com/guide/topics/ui/declaring

-layout.html.

Google Inc. (2012b). Google Maps Android API -

External Library. Retreived June 1, 2012, from

https://developers.google.com/maps/documentation/an

droid/index

Gramlich, N. (2012). OSMDroid: OpenStreetMap-Tools

for Android (Version 3.0.8) [Software]. Avaliable

from http://code.google.com/p/osmdroid/

Holt-Lunstad, J., Smith, T. B., & Layton, J. B., (2010).

Social relationships and mortality risk: A meta-

analytic review. PLoS Medicine, 7(7) doi:

10.1371/journal.pmed.1000316

Jarvinen, J., DeSalas, J. & LaMance, J., (2002). Assisted

GPS: A Low-Infrastructure Approach. Retrieved

March 1, 2002, from http://www.gpsworld.com/wp-

content/uploads/2012/09/gpsworld_Innovation_0302.p

McNair, D., (1984). Citation classic - manual for the

profile of mood states. Current Contents/social &

Behavioral Sciences, (27), 20-20.

Ohida, T., Kamal, A. M. M., Uchiyama, M., Kim, K.,

Takemura, S., Sone, T and Ishii, T., (2001) The

influence of lifestyle and health status factors on sleep

loss among the Japanese general population. SLEEP,

24(3), 333–338.

Ramos-Castro, J., Moreno, J., Miranda-Vidal, H., García-

González, M .A., Fernández-Chimeno, M., Rodas, G.

& Capdevila, Ll., (2012). Heart Rate Variability

analysis using a Seismocardiogram signal. In

Engineering in Medicine and Biology Society, EMBC

2012. Accepted.

Rodriguez-Ibanez, N., Garcia-Gonzalez, M. A.,

Fernandez-Chimeno, M., & Ramos-Castro, J., (2011).

Drowsiness detection by thoracic effort signal analysis

in real driving environments. 33rd Annual

International Conference of the IEEE Engineering in

Medicine and Biology Society, EMBS 2011, Boston,

MA. 6055-6058. doi: 10.1109/IEMBS.2011.6091496

Shin, J., Shin, D., Shin, D., Her, S., Kim, S., & Lee, M.,

(2010). Human movement detection algorithm using 3-

axis accelerometer sensor based on low-power

management scheme for mobile health care system

(Hualien ed.) doi: 10.1007/978-3-642-13067-0_12

Sposaro, F., & Tyson, G., (2009). iFall: An android

application for fall monitoring and response. 31st

Annual International Conference of the IEEE

Engineering in Medicine and Biology Society:

Engineering the Future of Biomedicine, EMBC 2009,

Minneapolis, MN. 6119-6122. doi:

10.1109/IEMBS.2009.5334912

Thomée, S., Härenstam, A. & Hagberg, M., (2011).

Mobile phone use and stress, sleep disturbances, and

symptoms of depression among young adults - A

prospective cohort study. BMC Public Health, 11 doi:

10.1186/1471-2458-11-66

Yamamoto, Y., (2011). Twitter4j: an open-sourced Java

library for the Twitter API. (Version 2.0.4) Avaliable

from http://twitter4j.org/en/index.html

Yang, S. & Gerla, M., (2011). Personal gateway in mobile

health monitoring. 2011 9th IEEE International

Conference on Pervasive Computing and

Communications Workshops, PERCOM Workshops

2011, Seattle, WA. 636-641. doi:

10.1109/PERCOMW.2011.5766966

Yuanchao, M., Bin, X. Yin, B., Guodong, S. and Run,

Zhu, (2012). Daily Mood Assessment Based on

Mobile Phone Sensing. In Wearable and Implantable

Body Sensor Networks, 2012 Ninth International

Conference on, 142-147.

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

248