driver warning assistant for monitoring heart rate and spo2 using mobile phones

8/9/2019 Driver Warning Assistant for Monitoring Heart Rate and SpO2 Using Mobile Phones

http://slidepdf.com/reader/full/driver-warning-assistant-for-monitoring-heart-rate-and-spo2-using-mobile-phones 1/8

Driver Warning Assistant for Monitoring Heart Rate and SpO2 usingMobile Phones

DUMITRU Adrian Iulian1, a *, MOGAN Gheorghe Leonte2, b

1Transilvania University of Brașov, B-dul Eroilor, 29 , Brașov, Romania

2Transilvania University of Brașov, B-dul Eroilor, 29 , Brașov, Romania

[email protected], [email protected]

Key words: heart rate, SpO2, driver, smart phone, pulse oximeter, android system, Bluetoothcommunication protocol.

Abstract: Nowadays both mobile phones and vehicles are playing an important role in our lives,

because these two industries have met an amazing development during the recent years. Also,

steady progress has been made in the medical sensors field and the increasing capacity of wirelesscommunication has led to the emergence of new methods for health monitoring. This paper presents

a system for real-time monitoring of the driver's health. The proposed driver warning assistant is

composed of a wearable low-power sensor that dynamically monitors the vital parameters of the

driver and the measured data are sent via Bluetooth wireless communication protocol. Data

processing, data management and communication and the human computing interfacing are

implemented using a smart phone running an Android operation system. Experimental results

associated with physiological parameters which have been measured are also included in the paper.

1. Introduction

Smart Phones are one of the most important devices in our lives and they are becoming verysuccessful in the area of health monitoring [1]. Due to the technological evolution hardware

components have been miniaturized and the mobile phones have become a kind of minicomputers

(smart phones). The smart phones of our days, besides the role of being mobile, have a robust

operating system (e.g. Android, iOS, Bada, BlackBerry, Windows Mobile, etc), a high power data

processing, Internet connectivity through different wireless communication techniques (GSM, Wi-

Fi, Bluetooth or other), or various integrated sensors such as camera, GPS, microphone, etc.

Because the potential of smart phones has been explored in many medical telemonitoring

applications and together with all the factors mentioned above, smart phones are an ideal option as a

"take anywhere" physiological monitor without the need for additional hardware [2, 3, 4, 5].

Driver’s health status is a very important factor regarding traffic safety. The monitored

parameters from this paper are the heart rate and the blood oxygen saturation. The limits of these parameters must be very well defined because any abnormal situation may lead to a potential traffic

accident. An Android platform based on Java was used to create the application for monitoring,

transmitting and storing the parameters discussed above [6].

The remainder of this paper is structured in 6 different sections summarized as follows: in

the second section, we discuss the related work. In section 3, we discuss the hardware system

architecture where we present a short description of the global block diagram. In section 4, we

present the software implementation of the application. The experimental test and results are

reported in section 5. Finally, the paper ends with a concluding section, section 6, where we also

make claims about the future work on this subject.

2. Related work

The development of physiological measurement systems for the automotive industry has

different incentives. Providing medical care and the functionality of personal health monitoring is a

mailto:[email protected]








motivation, especially in the context of demographic changes and aging drivers. Another reason for

developing applications using biomedical instruments is the desire to use this new human-machine

interface (HMI) as a channel of information on the health of the driver [7].

A methodology to continuously monitor the heart rate variability and the hemodynamic

stability of a car driver is essential, both for the driver and the passengers of the vehicle. Emarose et

al. [7] discusses the maximum physiological parameters which can measure by means of anoximeter installed on the steering wheel of the car. The oximeter which is normally used to measure

the oxygen saturation from the blood is exposed to the magnetic fields in order to obtain

information about the blood viscosity, a parameter that is essential to keep the body stable from a

hemodynamic point of view.

In case of a heart failure, the vehicle is changed to the Cruise Control module and the

measured parameters will be sent via GPS to any nearby hospital. The RFID tags are also used to

communicate the state of emergency to the vehicles nearby. Each vehicle is equipped with an

automatic external defibrillator (AED) with detailed instructions on how to operate the AED for the

defibrillation of the person suffering the heart failure.

For the monitoring of the driver, Geva et al. [8] proposed a system that includes several

sensors to monitor the vital signs, connected or integrated at least one element of the vehicle. Thesesensors are for monitoring the ECG, the heart rate, the oxygen saturation, the temperature and the

blood pressure, measuring driver ’s breathing, glucose, weight or environmental monitoring.

Gabaude et al. [9] examine the relationship between mental effort and performance

management by making three different measurements: measuring the cardiac activity, the driving

performance and the subjective data. Using empirical research, this study attempts to understand

how the drivers manage their mental effort while driving in a simulator. So, the measurement of

driver’s mental effort is investigated by manipulating the road type and the performance of two

extra tasks (it is possible that some drivers can have a different mental effort than others). The

cardiac response pattern for the mental effort made during the act of driving a car has been

characterized by an increase in the heart rate and a decrease in the heart rate variability, when the

mental effort was increased and vice versa.

Riener et al. [10] have investigated through the measured ECG and GPS data collection the

application of the heart rate variability (HRV) analysis to electrocardiography (ECG) data for

identifying the driving situations of a possible threat by monitoring and recording the autonomic

arousal states of the driver.

Sun et al. [11] have developed a sensor that can detect ECG signal from a distance of 20 cm

to 30 cm away through cloth. The ECG sensor used is a non-contact one and he detects the potential

of the human body caused by capacitive coupling neuronal activity. Such high sensitivity makes it possible for practical implementation for driver physiological signal monitoring purpose.

Nowadays, Android based on Java is the most famous platform of smart devices, for

example, smart phone and PC tablet, including Android SDK [12]. Scully el al. [3] shows that amobile phone can serve as an accurate monitor for several physiological variables, based on its

ability to record and analyze the varying color signals of a fingertip placed in contact with its

optical sensor. They confirm the accuracy of measurements of breathing rate, cardiac R-R intervals,

and blood oxygen saturation, by comparisons to standard methods for making such measurements

(respiration belts, ECGs, and pulse-oximeters, respectively).

3. Hardware system architecture

To make the wireless connection between sensors, smart devices and automotive system,

Bluetooth is used as a transport protocol. The application is implemented on an Android platform

that includes the graphical user interface (GUI). The application runs on any Android devices, for

example, the smart devices or automotive systems. [12].

Fig.1 shows the general hardware architecture of the proposed system. The driver will put

the monitoring device (pulse oximeter) around his hand or his earlobe, and the device will start

recording the values. The data from the pulse oximeter device are wirelessly transmitted using a



Bluetooth connection to the smart phone which includes the application for communication, data

storage and user interface developed under Android OS. So, the user will be able to view the

measured values in real time and when he decides to close the application, the data will be saved in

a .csv file named DriverLife(0…n) (where “n” is the last saved file). In a future work, when running

the application, if it is found that one of the vital signs monitored exceeded the normal value,

initially the system will be able to warn the driver to stop the car. If it is determined that the driveris unable to be still driving, the system will be able to intervene and safely stop the vehicle.

Fig. 1 Hardware block diagram

The pulse oximeter is a non-invasive sensor that measures oxygen saturation by shininginfrared and red light through apart of the body, such as a finger, toe or ear. The non-invasiveness

and potential low cost of this sensor technology facilitates its mobility and use in low resource

settings [13]. The sensor measures the amount of red and infrared light received by the detector and

calculates the amount absorbed. Much of it is absorbed by the tissue, bone and venous blood, but

these amounts do not change dramatically over short periods of time [14]. In other words the

amount of arterial blood does change over short periods of time due to pulsation (although there is

some constant level of arterial blood). Because the arterial blood is usually the only light absorbing

component which is changing over short periods of time, it can be isolated from the other

components [5].

4. Software system architecture

Fig. 2 shows the graphical interface of the application running on the smart phone. The main

function of this application is to measure the heart rate and the SpO2. The "Heart Rate" parameter

indicates that the heart rate is detected, measured in bpm (beats per minute) (the red color in the

graph of Fig 2), and the "SpO2 (%)" parameter is the blood oxygen saturation measured as a

percentage (the blue color in the graph of Fig. 2).

Fig. 2 The graphical interface of the application on start and in running

The following parameter named “Pleth” refers to the measurement of plethysmography

(which records changes in the volume of the anatomical segment relative to the systolic-diastolic

scheme blood flow through that segment) and it is represented in the graph from Fig. 2 in grey. The

data from the “Level” parameter builds the graph from the bottom of the Fig. 2 (yellow color)

which highlights how strong the pulse at each heartbeat is.

The “Signal Strength” parameter refers to the signal level of connectivity between the two

devices (smart phone - oximeter) and the “Pulse Alert” prevents the driver if he is in a bad mood for



Data updating diagram on GUI. The graphical user interface (Activity) from the Android

operating system has a structure similar to a "tail", where the other thread scan places different

activities ("Runnable") to a free update without blocking it.

Fig. 5 Flow data updating diagram

The activity placed by the acquisition thread includes the relative time point to which the

package was sampled and all the data derived from there. This activity will invoke the graphical

interface method "UPDATE UI", forwarding its embedded data to it.

If the data received is invalid (null), it is a sign that the graphical interface must be set to default.

Otherwise, the data contained are used to update values and graph (plot) of the graphical interface

and some of them are placed in the file.

5. Experimental tests and results

The pulse oximeter measures the blood-oxygen saturation levels (SpO2) as well as the heart

rate. The measurements are based on Lambert Beer’s law of spectral analysis which relates theconcentration of absorbent in solution to the amount of light transmitted through the solution [15].

In Fig. 6 an example with all the measured parameters is shown: the heart rate, the spo2, the level

and the pleth.

Fig. 6 Measured parameters

The measurements were performed on two subjects. They drove the car both in urban and in

rural areas. In the urban area the measurements were made at peak hours and when the traffic was

relatively small. In most cases the monitored parameters remained constant, slightly higher in urban

areas, in crowded situations.

The primary function of the heart is to supply blood and nutrients to the body. The regular

beating or contraction of the heart moves the blood throughout the body. A regular heart beat of a

heart has a frequency between 60-100 beats per minute; it seems however that these frequency

limits are not exactly correctly chosen: on the one hand, there are plenty of healthy individuals

whose heart rate is 50-60 beats per minute, and on the other hand, it was found that standing heart

rates between 85-100 beats per minute are neither signs nor prerequisites for a full cardiac health.

Therefore, according to the latest research, normal heart rate should rather be defined between 50-

85 beats per minute [16, 17].



In Fig. 7(a) we calculated the frequency of occurrence of the pulse and the normal

distribution (Gaussian) for the first subject monitored. Thus it can be seen that most values are in

the range of 85-95 beats per minute. In Matlab we built a script that imports all files. csv saved in

the smart phone and performs the graph in the Fig. 7 and calculates the standard deviation equals to

the value 5.7454 for the first subject, and the value 3.1535 for the second one. For the second

subject, there was a lower heart rate that fits in the range of 60-80 beats per minute.

Fig. 7 Subject 1- Heart rate occurrence (a); Subject 2 - Heart rate occurrence (b)

Comparing the two graphs with the measured data, it can be seen that the pulse rate is

constant for both subjects. In Fig. 8(a) it is presented a heart rate comparison of these two healthy

subjects and beside the fact that a minor difference between this two heart rates exists, both are

considered normal because each individual’s pulse varies from case to case. In these conditions, in

future works, one will make a database with the data measured for example for a week and calibrate

the system customized for each user. Thus, if a person has a higher or a lower pulse, but it is

perfectly healthy from a medical standpoint, the system does not warn the user if it is registering

higher or lower values than normal.

Fig. 8 Heart rate comparison (a) Heart rate comparison - simulation (b)

In Figure 8 (b) a situation in which the heart rate falls below normal levels is simulated,reaching 25 to 30 bpm. In these circumstances, the driver may lose consciousness due to

insufficient irrigation of the specific tissue of the myocardium, which can be fatal.

For a regular healthy person, the normal blood oxygen saturation level (SpO2) should be

around 94% to 99%. For peoples with mild respiratory diseases, the SpO2 should be 90% or above.

Supplementary oxygen should be used if SpO2 level falls below 90%, which is unacceptable for a

prolonged period of time. Basically, a saturation of 97% of the total amount of hemoglobin in the

body is filled with oxygen molecules [18, 19]. Regarding SpO2 measurement, the results showed

that the measurements are within the normal parameters, where most values revolve around 97%, as

can be seen in Fig. 9.

As seen, the duration of the data measurement (measured in percent) did not decrease below

the value of 94%. This leads to the conclusion that these subjects do not have respiratory problems,

resulting a normal amount of oxygen concentration in the blood.



Fig. 9 Spo2 measurement comparisons

6. Conclusions and future work

This paper presents the structure of a driver assistant implemented as an android mobile

application designed to measures in real time the heart rate and SpO2 parameters of a driver and

sonor or/and light warning the driver in case of inadequate values of these parameters. The

connection between the smart phone and the pulse oximeter was made through Bluetooth

communication protocol. All the data captured were saved on the smart phone's memory and after

this they were processed.

We chose the pulse-oximeter for measure our interest parameters because he has become

one of the most common physiological monitors used today in hospitals. Moreover, algorithms have

been reported to detect atrial fibrillation, blood loss, and autonomic nervous system disorders, in

addition to traditional vital sign measurements of the heart rate, the respiration rate, and the oxygen

saturation from the dynamics in a pulse-oximeter signal.

One major advantage for the driver is that he can see every moment his recording

parameters on the graphical user interface of his phone. This driver warning assistant can beintegrated in other mobile applications regarding face and eye detection, breath rate or body

temperature. Personal healthcare services on automotive environment can be helpful for elderly

people and chronic diseases. It can prevent terrible accidents due to some health problems of the

driver.

Experimental test results were encouraging, because the measured physiological parameters

of the two subjects who were perfectly healthy were rated in the normal range of health.

In a future work, we will develop a based data and a package to personalise the application

for each individual user. The user in question will calibrate the system according to its own

parameters.

Because modern cars now have more and more equipped with functionalities dependent on

embedded electronics [20], we will try to integrate this assistant car breaking and/or steeringsubsystems. So we are trying to create a robust system for monitoring the driver health and act on

car subsystems in order to avoid as many accidents as possible.

7. Acknowledgements

This work was partially supported by the strategic grant POSDRU/159/1.5/S/137070 (2014)

of the Ministry of National Education, Romania, co-financed by the European Social Fund –

Investing in People, within the Sectoral Operational Programme Human Resources Development

2007-2013.

8. References

[1] G. H. Nanhore and M. M Bartere: Mobile Phone Sensing System for Health Monitoring,

International Journal of Science and Research (IJSR). Volume 2, Issue 4 (2013), pp. 252-255.



[2] D. Barata, G. Louzada, A. Carreiro and A Damasceno: System of Acquisition Transmission

Storage and Visualization of Pulse Oximeter and ECG Data Using Android and MQTT, Procedia

Technology. Volume 9, (2013), pp. 1265-1272.

[3] C. G. Scully, J. Lee, J. Meyer, A.M. Gorbach, D. Granquist-Fraser, Y. Medelson and K.H.

Cho: Physiological Parameter Monitoring from Optical Recordings with a Mobile Phone, IEEE

Transaction Biomedical Engineering, Volume 59, Issue 2 (2012), pp. 303-306[4] O. Postolache, P.S. Girão, M. Ribeiro, M. Guerra, J. Pincho, F. Santiago and A. Pena: IEEE

International Workshop on Enabling Telecare Assessment with Pervasive Sensing and Android OS

Smartphone, in Medical Measurements and Applications Procedings (MeMeA)(2011), pp. 288-293.

[5] S.J. Jung, R. Myllylä and W.Y. Chung: Wireless Machine-to-Machine Healthcare Solution

using Android Mobile Devices Global Networks, IEEE Sensor Journal. Volume 13, Issue 5 (2013)

pp. 1419-1424.

[6] E. Burnette, Hello, Android: Introducing Google's Mobile Development Platform

(Pragmatic Programmers), Pragmatic Bookshelf Ed., Nov. 2009.

[7] S. Emarose, R. Asokan and D. Valayaputtur: Continuos Monitoring Heart Rate Variability

and Heamondynamic Stability of an Automobile Driver to Prevent Road Accidents, Third

International Conference on Computing Communication & Networking Technologies (ICCCNT),(2012) pp. 1-7.

[8] N. Geva, Y. Geva and Y. Tal, U.S. Patent 20,130,070,043. (2013)

[9] C. Gabaude, B. Baracat, C. Jallais, M. Bonnaiaud and A. Fort: Cognitive Load Measurement

While Driving, HFES Europe Chapter Conference Toulouse, 2007.

[10] A. Riner, A. Ferscha and M. Aly: Heart on the Road: HRV Analysis for Monitoring a

Driver’s Afective State, in Proceedings of the 1st International Conference on Automotive User

Interfaces and Interactive Vehicular Applications, (2009), pp. 99-106.

[11] Y. Sun, X. Yu, J. Berilla, Z. Liu and G. Wu: An in-Vehicle Physiological Signal Monitoring

Sistem for Driver Fatigue Detection, Submitted to the 3rd International Conference on Road Safety

and Simulation (2011).

[12] K. Han, M. Jung and J. Cho: Implementation of the Personal Health Care Service on

Automotive Environments, Personal and Ubiquitous Computing, Volume. 18, Issue 3 (2014), pp.

523-533.

[13] D. Parekh: Design Heart Rate Blood Pressure and Body Temperature Sensors for Mobile

on-Call Sistem, , PhDr thesis, Department of Electrical and Computer Engineering, McMaster

University (2010).

[14] S. Behbahani and M. A. Pishbin: New Oxygenation Method Based on Pulse Oximeter,

American Journal of Biomedical Engineering, Volume 2, Issue 4 (2012) pp.185-188.[15] O. O. Ogunduyile, O.O. Oludayo and M. Lall: Healthcare Monitoring System Using a

Collection of Sensor Nodes, International Journal of Advanced Research in Computer Science &

Technology, Volume 3, Issue 2 (2013), pp.632-639.[16] J. Achten and A. E. Jeukendrup: Heart Rate Monitoring, Sports Medicine, Volume 33, Issue

7 (2003), pp. 517-538.

[17] P. Palatini: Need for a Revision of the Normal Limits of Resting Heart Rate, Hypertension

(1999).

[18] M. Shafique, P. A. Kyriacou and S. K. Pal: Investigation of Photoplethysmographic Signal

and Blood Oxygen Saturation Value in Healthy Volunteers During Cuff-Induced Hypoperfusion

Using a Multimode PPG/SpO2 Sensor, in Medical & Biological Engineering & Computing, Volume

50, Issue 6 (2012), pp. 575-583.

[19] S. L. Schultz: Oxygen saturation monitoring by pulse oximetry, AACN procedure manual

for critical care (2001).

[20] J. L. Boulanger and Q. Ð. Van: Requirements Engineering in a Model-based Methodologyfor Embedded Automotive Software, IEEE International Conference on Research, Innovation and

vision for Future (2008), pp. 263-268.

http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=6383183


http://link.springer.com/journal/40279

http://link.springer.com/journal/40279/33/7/page/1





http://link.springer.com/journal/40279




driver warning assistant for monitoring heart rate and spo2 using mobile phones

Documents