Health status and air quality parameters monitoring based on ... - IPS

Int. J. Advanced Media and Communication, Vol. 3, Nos. 1/2, 2009

Health status and air quality parameters monitoring based on mobile technology and WPAN O.A. Postolache* and P.M.B. Silva Girão Instituto de Telecomunicações/DEEC-IST, Av. Rovisco Pais, 1049-001, Lisboa, Portugal E-mail: [email protected] E-mail: [email protected] *Corresponding author

P. Sinha and A. Anand Indian Institute of Technology, Guwahati, India E-mail: [email protected] E-mail: [email protected]

G. Postolache Escola Superior de Saúde, Universidade Atlantica, Antiga Fábrica da Pólvora de Barcarena, 2730-036 Barcarena, Portugal E-mail: [email protected] Abstract: The paper presents a Wireless Personal Area Network (WPAN) including two Bluetooth-enabled measuring nodes that delivers information about some physiological parameters extracted from photo-plethysmographic signals and provides information of indoor air temperature and relative humidity. The data from the node of physiological and air parameters measurement are received by a smart phone device through Bluetooth communication. J2ME developed software assures Bluetooth device detection and identification, data reading, primary processing, data storage and automatic alarm generation. Critical health status associated with the assessed person leads to an automatic SMS generation. Elements related to Heart-Rate Variability (HRV) estimation are also included in the paper. This is an expanded version of a paper presented at the 3rd IEEE International Workshop on Medical Measurements and Applications, 9–10 May 2008, Ottawa, ON, Canada. Keywords: smart sensors; homeostasis monitoring; personal wireless network; smart phone GUI; mobile data processing. Reference to this paper should be made as follows: Postolache, O., Silva Girão, P., Sinha, P., Anand, A. and Postolache, G. (2009) ‘Health status and air quality parameters monitoring based on mobile technology and WPAN’, Int. J. Advanced Media and Communication, Vol. 3, Nos. 1/2, pp.139–153.

Copyright © 2009 Inderscience Enterprises Ltd.

139

140

O.A. Postolache et al. Biographical notes: Octavian A. Postolache received the PhD Degree in Electrical Engineering from the Faculty of Electrical Engineering, Technical University of Iasi, Iasi, Romania, in 1999. He worked as an Assistant Professor with Faculty of Electrical Engineering, Technical University of Iasi, and from 2000, he started working as a PhD Researcher with the Instituto de Telecomunicações Lisbon, Portugal, where he was involved in different projects in the area of instrumentation. His main research interests concern smart sensors for environment and biomedical applications, instrumentation networks, and computational intelligence implementation in automated measurement systems. P.M.B. Silva Girão received the PhD Degree in Electrical Engineering from the Instituto Superior Técnico of the Technical University of Lisbon (IST/UTL), Lisbon, Portugal, in 1988. In 1975, he joined the Department of Electrical Engineering, IST/UTL, first as a Lecturer and, since 2007, as a full Professor. Presently, his main research interests are instrumentation, transducers, measurement techniques and digital data processing particularly for biomedical, environmental, chemical, and civil applications. Metrology, quality, and electromagnetic compatibility are also areas of regular activity mainly as auditor for the Portuguese Institute for Quality (IPQ). Priyesh Sinha studied at Indian Institute of Technology, Guwahati, India, from 2004 to 2008. He received his BTech Degree in Mechanical Engineering from the Indian Institute of Technology, Guwahati, India, in 2008. He is currently working as a Mechanical Designer in Shell Technology India based in Bangalore. Abhinav Anand studied at Indian Institute of Technology, Guwahati, India, from 2004 to 2008. He received his BTech Degree from the Indian Institute of Technology, Guwahati, India, in 2008. He is currently Software Engineer in Samsung India Software Operations, Bangalore, India working in the development of Technology in High Speed Packet Access (HSPA). Gabriela Postolache graduated in Biology at Al.I. Cuza University, Iasi, Romania in 1992. She is specialised in Enzimology and Physiology and holder of a PhD in Physiology. Presently, she is Professor of Human Physiology and Cell and Molecular Biology in the Superior School of Health of the Atlântic University. Her most important research was carried out in the field of the autonomic nervous system analysis. Her research interests broadly surround cardiovascular and respiratory physiology, heart rate variability and blood pressure variability, physiopathological effects of alcohol and caffeine on organisms, electrophysiology, biomedical measurements and Home TeleCare.

1

Introduction

It is well recognised that many factors contribute to health and illness, such as socioeconomic conditions, individual lifestyles and behaviours, culture, genetics and access to health services. Environmental factors can have a positive or negative influence on a person’s participation as a member of society, on performance of activities, or on a person’s body function or structure. The new international classification of health – International Classification of Functioning, Disability and Health ICFDH

Health status and air quality parameters

141

(World Health Organization, 2001), besides body functions and structures, activities, and participation, includes also a list of environmental factors in the classification of health and health-related domain. ICFDH designates as environmental factors: products and technology, natural environment and human-made changes to environment, support and relationship, attitudes, services, systems and policies. Health Information Technology (HIT), including Electronic Medical Records (EMRs) and their separate functionalities, is an important tool for services associated with health care. EMRs help the individual to maintain their normal health profile by providing better monitoring and feedback so that the earliest signs of health problems can be detected and corrected. The large benefits found in the improvement of fundamental aspects of patient care (Rind et al., 1994; Petersen et al., 1998; Kuperman et al., 1997) indicate that information technology can be an important tool for improving safety in many clinical settings. The recent literature on the effect of HIT on the quality and efficiency of care shows that EMRs system can increase the delivery of the care, enhance the capacity of the providers of healthcare to perform surveillance and monitoring for disease conditions and care delivery, reduce rates of medications errors, and decrease utilisation of care (Chaudhry et al., 2006). This can be accomplished at affordable cost by collecting a wide range of vital signals and relevant data about patients, providing early monitoring and warning systems for people with high-risk medical problems, supplying that information to providers on request, permitting physicians to enter patient care orders on the computer, and providing health professional with advice for making healthcare decisions about individual patients. A recent survey (Schoen et al., 2006) involving 6000 physicians from Australia, Canada, Germany, the Netherlands, New Zealand, the UK and the USA showed a widespread distribution of EMRs in the Netherlands (98%), New Zealand (92%), the UK (89%) and Australia (79%) comparative with USA (28%) and Canada (23%). Use of computerised alerts among physicians ranged from 93% (the Netherlands) to 10 % in Canada and 23% in USA. A study on potential savings and costs of widespread adoption of EMRs systems in the USA concludes that effective EMRs implementation and networking could eventually save more than $81 billion annually – by improving healthcare efficiency and safety – and that HIT enabled prevention and management of chronic disease could eventually double those savings at the same time increasing health and other social benefits (Hillestad et al., 2005). The investigators estimated that achieving a 90% rate of adoption of EMRs in hospital would cost additional $121 billions over a period of 15 years but would yield net savings of $531 billions over the same period. However, this is unlikely to be realised without related changes in the healthcare system. The wide availability of high-bandwidth public wireless networks has given rise to new mobile healthcare services (Konstantas et al., 2004). Referring to homeostasis monitoring, the temperature, the heart rate, the respiration rate and the blood pressure are reported as the main homeostasis quantities (Boychuk et al., 2006; von Kanel et al., 2005; Task Force of the European Society of Cardiology and The North American Society of Pacing and Electrophysiology, 1996) and different commercial systems are used to provide information about these parameters. The implemented architectures encompass the ability to collect multiple types of data from both static and mobile sensing units. An important feature of the distributed healthcare monitor architectures is the inclusion of both wired and wireless communication protocols. Thus, the static sensing units are generally characterised by wired communication protocols such as RS232, USB2.0 or IEEE802.3, whereas the wireless communication protocols are

142

O.A. Postolache et al.

expressed by IEEE802.11a/b/g, IEEE802.15.1 or IEEE802.15.4 (Rahman et al., 2005; Gao et al., 2007; Chan et al., 2008). To join static and mobile solutions, wired-to-wireless bridges are used. These bridges make possible to transfer homeostasis data or environment data over either a homogeneous or hybrid wireless network to a central server, and perform data fusion with computer-based decision support to extract vital information about the health status of the person being monitored and the environmental condition where he or she lives. Beyond the principal improvement in management of a resident’s disease and wellness, the major benefits of the system we propose include its ubiquitous and unobtrusive nature and an elevated objectivity in performing functional health assessment together with environment condition assessment using low-cost and high autonomy wireless sensing units and commonly used devices such as a 3G smart phones to materialise the Human–Machine Interface (HMI). Referring to homeostasis monitors and corresponding HMI, portable solutions based on Notebooks (Bellazzi et al., 2001; Postolache et al., 2007) and mobile solutions based on Personal Digital Assistants (PDAs) (Gao et al., 2008; Postolache et al., 2006) are the most common used solutions. In the Notebook-based solution, the increased capability of signal processing together with the HMI functionality makes the solution still interesting when complex evaluation of the health condition, which involves intensive data processing of high amount of acquired data is required. With increased mobility but with important restrictions at the computation times and data storage levels, PDAs are nowadays the main solution for HMI and primary data processing associated with healthcare monitoring systems. However, PDAs are more expensive than 3G smart mobile phones, which additionally have different kinds of wireless communication interfaces (e.g., WiFi, Bluetooth). Mobile phone based solutions are mentioned as possible interfaces with distributed measurement systems from the earliest of 2000. However, in most of the applications the mobile phones use the SMS service to send and receive commands or data (Gómez et al., 2002). In recent years, mobile phones running Symbian OS (Nokia, Motorola and Sony-Ericson) create the possibility to develop mobile healthcare software (de Jode, 2005; Sybase, 2008). The challenge of the work now reported was to develop a flexible and low-cost solution for mobile healthcare. Thus, we implemented and tested a distributed hybrid measurement system that includes static and mobile measurement nodes wireless connected and a HMI expressed by a 3G mobile smart phone. The system performs acquisition of the information from physiological measuring nodes as well as from indoor air quality measuring nodes correlating the patients’ health status and indoor air conditions (temperature and humidity). A physiological measuring node consists of an embedded photoplethysmograph based on a Microchip PIC18F452 microcontroller and RS232-Bluetooth bridge. Health status parameters such as heart rate, Heart-Rate Variability (HRV), and saturation of oxygen in arterial blood flow are calculated at the microcontroller level. The indoor air quality measuring nodes are implemented using data acquisition Bluetooth-enabled units connected to the relative humidity and temperature measuring channels. An important part of the work is related to Java2ME programming of a 3G smart phone to assure data reading from the embedded optical plethysmograph and air quality measuring nodes through the Bluetooth communication. Additionally, the mobile device assures the voltage to air quality parameter value conversion, alarm signal generation according to the physiological parameters status and imposed thresholds, and automatic SMS generation for the critical patient health conditions.


143

Different communication tests were implemented using an additional Notebook that is also used for advanced signal processing of HRV using wavelets (Newandee and Reisman, 2003), data logging and data publishing.

2

WPAN for physiological and air quality parameters measurement

The Wireless Personal Area Network (WPAN) designed for health status and indoor air conditions assessment is a Bluetooth piconet network that includes a physiological measuring node (Phy) and single or multiple indoor air quality nodes (AirQ) as smart phone piconet slaves (Figure 1). Figure 1

Physiological parameters and air quality sensing network topology (RHS – relative humidity sensors; TS – temperature sensors; m/s – master/slave) (see online version for colours)

The piconet master (m/s) is a 3G Smart Phone that works as a user interface (HMI node) with the Bluetooth-enabled devices and performs the data storage and data processing of the voltage values received from the air quality nodes. Considering the limited capabilities of the smart phone regarding data processing and data storage, a laptop PC Bluetooth compatible is included in the distributed sensing system and works as master of the piconet laptop PC that includes the Smart Phone as a piconet slave. The laptop PC performs advanced data processing, data logging and data publishing as well as the sensing nodes testing during the network implementation.

2.1 Physiological parameters measuring node Considering the acquisition, primary processing and communication requirements associated with the WPAN’s physiological parameters measuring node, an embedded solution based on the PIC18F452 microcontroller was designed and implemented. Control and acquisition tasks implemented at the microcontroller level are associated

144


with the developed Two-Channel Optical Plethysmograph (2ch-OP). The 2ch-OP is characterised by the following conditioning circuit blocks: current control of two optical sources associated with a two-channel optical plethysmograph, infrared (IR) and red light detection block, amplification and filtering block (Figure 2). Figure 2

Physiological parameters measuring node block diagram (CC – current driver digital control)

After signal conditioning, the voltage signals obtained from the IR and red measuring channels are acquired using two analogue input channels of the PIC18F452 microcontroller. The development of a two-channel optical plethysmograph is related to the differential absorption measurement of red and infrared light by blood’s haemoglobin and oxyhaemoglobin. The differential absorption is measured using two optical detectors, TSL257 and TSL260 light-to-voltage converters placed in a finger in the opposite side of IR and red light LEDs. The TSL257 and TSL260 are high-sensitivity low-noise light-to-voltage optical converters that combine a photodiode and a transimpedance amplifier on a single monolithic CMOS integrated circuit. The output voltage is directly proportional to light intensity (irradiance) on the photodiode. The irradiance responsivity is 1.68 V/(W/cm2) for λ = 645 nm for TSL257 and 0.042 V/(W/cm2) for λ = 940 nm in the TSL260 case. The devices have improved offset voltage stability, low power consumption and accept single power supply (5V). The excitation LEDs of the plethysmograph-sensing probe are controlled using TTL pulse signals that are generated by using the programmed RD0 and RD1 digital outputs of the PIC18F452 microcontroller. To vary the IR and red LEDs excitation currents, a digital potentiometer scheme based on Xicor X9C104 as part of a current driver was designed and implemented. The digital potentiometer control is performed using three


145

digital lines (RD2, RD3 and RD4) of the microcontroller. The timing signals for the LED control are generated using TIMER0 interrupt. Each LED (red and IR) is on for 0.25 ms and off for 0.75 ms giving a total duration of 1 ms and a duty cycle of 25%. During the LED excitation period, the voltages from the optical detectors after filtering (PPG ac and dc components) are acquired using the RA0, RA1, RA2 and RA5 analogue inputs of the microcontroller. The band pass filter block with pass band of 0.025–10 Hz is obtained as a sequence of a high-pass and low-pass active filters applied to the red and IR detectors as only the frequencies within this range are required according to the lowest and highest values of human heart rate (Li, 2007). The time stamp and the voltage values acquired from the optical detectors (photoplethysmography samples) are saved in the mobile device memory and are used to calculate the oxygen saturation in arterial blood. The arterial blood oxygen saturation (SpO2) requires the implementation on the mobile device of a virtual pulse oximeter software component (Kyriacou1 et al., 2007). The virtual pulse oximeter uses the ac component of the photoplethysmographic (PPG) signal (Figure 3) obtained after filtering using a set of analogue active filters as shown in Figure 2. The amplitudes of the red and infrared ac PPG signals are sensitive to changes in arterial oxygen saturation owing to the differences in the light absorption of oxygenated and deoxygenated haemoglobin at these two wavelengths (λred = 645 nm, λIR = 940 nm). From the ratios of these amplitudes the arterial blood oxygen saturation (SpO2) is estimated. Hence, the technique of pulse oximetry relies on the presence of adequate peripheral arterial pulsations, which are detected as PPG signals (Mendelson and Ochs, 1988). Figure 3

Red and infrared photoplethysmographic waves measured by transition through tissue

To estimate blood oxygen saturation (SpO2) a simplified ratio method was used. Thus, the ratio associated with SpO2 values is given by relation: VredPPGac R=

VirPPGac

VredPPGdc

(1)

VirPPGdc

where VredPPGac and VredPPGdc represent the acquired samples for λred = 645 nm and VirPPGac and VirPPGdc represent the acquired samples for λir = 940 nm. The acquired samples are sent to the mobile device to calculate the R value and to estimate the blood oxygen saturation using the SpO2 = SpO2experimental(R) experimental characteristic given in the

146


literature (Kamat, 2002). A polynomial model of the above-mentioned characteristics was computed and corresponds to the 75–100% SpO2 measurement range. The SpO2 curve vs. the calculated R ratio is presented in Figure 4(a) while the 3rd order polynomial model error evolution for the given operational range is presented in Figure 4(b). Figure 4

SpO2 experimental characteristics: (a) SpO2 dependence and (b) evolution of modelling error

(a)

(b)

The implemented relation in the Smart Phone software is: 3

SpO2(%) = ¦ an ⋅ R n

(2)

n=0

where, for the particular considered SpO2 experimental curve, a0 = 102.04%, a1 = 5.92%, a2 = −31.59% and a3 = 9.85%. Additionally, heart rate and HRV measurement modules are implemented at the microcontroller level. Two timers of the microcontroller are used. TIMER1 is used to detect the heart beat pulses corresponding to 5 s time interval and calculates the average value in 60 s. TIMER2 is used to count the time elapsed between occurrences of successive peaks. TIMER2 is started just after enabling TIMER1 and an external interrupt on pin PORTB.RB0 is turned on. Whenever an interrupt occurs, the TIMER2 count value is read and it is again initialised to 0 and starts counting again. HRV is estimated by subtracting the successive measured peak-to-peak time intervals. The variation in the time period separating consecutive heartbeats, conventionally described as HRV, was analysed using different techniques. Programming the PIC18F452’ UART (9600bps, 8bit, no parity, 1 stop bit, no flow control) and using Firefly D2D4 RS232-Bluetooth bridge (configured as DCE), the physiological measurement node is able to be read by a Bluetooth-enabled smart phone (Nokia N70 in the present case) or by the Bluetooth-enabled laptop PC for advanced data processing and data logging.


147

2.2 Indoor air parameters measuring node The indoor air conditions are measured during the physiological parameters measurement using a set of temperature (LM35) and relative humidity sensors (Honeywell HIH3610). Additionally, a non-inverter amplifier scheme was associated with the temperature measuring channel. The transfer characteristics of the temperature measurement channels associated with AirQ nodes are described by: VT (3) G ⋅ ST where G is the amplifier gain (G = 11), ST represents the temperature sensor sensitivity (ST = 10 mV/°C) and VT the voltage acquired from the temperature measurement channel. Referring to the relative humidity measuring channel, the temperature compensated RH value is given by the following relation: T ( o C) =

§ V · 0.062 × ¨ RH − 0.16 ¸ ¨V ¸ © supply ¹ RHT = 1.0546 − 0.00216 ⋅ T

(4)

where T is obtained from the temperature channel, VRH is the voltage acquired from relative humidity measurement channel and Vsupply represents the sensor voltage supply (Vsupply = +5V). The voltages from indoor air quality sensors nodes are applied to the analogue inputs Ain0, Ain1 of 16-bits Data Acquisition Module Bluetooth enabled (BlueSentry). Considering that the data acquisition module does not provide signal processing capabilities, the voltages acquired from the temperature and relative humidity channels are read by the smart phone in hexadecimal format and are afterwards converted into RHT values. The data acquisition module control provides a set of commands that are used to perform single-point and multiple-point data acquisitions.

3

Smart Phone embedded software

The development of applications for smart phones can be done on different platforms like embedded C (Khalifa and Novacky, 2006), J2ME (Kumar et al., 2004) and .NET, just to name a few. However, J2ME (Java2 microedition) gives ample options and is a rich platform to develop applications for smart phones. That is why we chose it. Additional software was also developed using Java APIs for Bluetooth (Ortiz, 2004). Different families of mobile devices have different memory capacity, user interfaces and processing power. J2ME specification allows choosing an appropriate Java environment for each ‘family’ of devices. This is done by selecting from J2ME configurations and profiles expressed by Connected Limited Device Configuration (CLDC). The CLDC defines the base set of application programming interfaces and a virtual machine for resource-constrained devices like mobile phones, pagers and mainstream personal digital assistants. The used Java APIs for Bluetooth (JSR-82) wireless technology (JABWT) is designed to operate under CLDC configuration. A J2ME profile adds further classes

148


to a configuration to produce a complete Java environment for a particular set of devices. The JSR-82 API hides the complexity of the Bluetooth protocol stack by exposing a simple set of Java API’s. The used classes in the present application were: BluetoothStateException, DataElement, DeviceClass, DiscoveryAgent, LocalDevice, RemoteDevice, and ServiceRecord. To develop the software, the Java Virtual Machine (JDK 1.6.0_01), Java Runtime Environment (JRE 1.6.0_01) and Java API for Bluetooth wireless technology (JSR-82) were installed in the laptop PC. The Netbeans IDE (Netbeans Mobility-5_5_1-windows) was also used for simulation of programs in the specific mobile phone model or its emulator platform. The developed application uses client–server architecture. The client software, which is implemented in the smart phone, provides the initialisation of the device, identification of the Bluetooth-enabled measuring nodes, data reading, data processing and alarm generation for patient health risk states. The flowchart associated with the developed application is presented in Figure 5. In the flowchart, the Serial Port Profile (SPP) defines the requirements for Bluetooth devices necessary to emulate serial cable connections using RF communication between two peer devices such as the smart phone and the measuring node, either the physiological parameter node or the indoor air quality parameter nodes. The requirements are expressed in terms of services provided to applications and by defining the features and procedures that are required for interoperability between Bluetooth devices. To assure higher interoperability, the critical health-state detector implemented in software assures the SMS automatic generation for an emergency phone number. Figure 5

The block diagram of the Bluetooth WPAN for physiological and indoor air quality measurement


4

149

Heart-Rate Variability

There is a substantial variance in HRV in normal individuals (Task Force of the European Society of Cardiology and The North American Society of Pacing and Electrophysiology, 1996) related mainly with differences in the parasympathetic control of heart and autonomic nervous system control of the cardiovascular system. HRV analysis has established itself as a non-invasive research and clinical tool for indirectly investigating both cardiac and autonomic system function in physiological condition and disease. The current methodologies used to quantify HRV are largely based on linear techniques that analyse electrocardiogram signal in the time and frequency domains. The present work investigates the utilisation of the time difference between the maximums of the plethysmography signal (Postolache et al., 2006) as a measurement of cardiovascular pulsations caused by the blood volume changes in the body. Considering computational load limitations, the HRV analysis technique implemented on the smart phone is expressed by the heart-rate standard deviation (σHR) of the time interval between consecutive peak-to-peak plethysmography (PPP) and root mean square of the difference between two adjacent PPP-intervals.

5

Results and discussions

The home health status monitor developed as a Bluetooth WPAN including the smart phone programmed in Java2ME performs the heart-rate measurement, oxygen saturation and beat-to-beat time interval variation as well as the indoor air temperature and relative humidity measurement. Different smart phone panels were developed considering the display resolution restriction (Figure 6). Regarding the implemented HMI, different panels correspond to different services. One of the first panels is the service detection panel that is used for the service search (Figure 6(a)). A second panel presents the service menu (Figure 6(b)). The user can select one of the services that are associated with the Bluetooth connection between the smart phone and physiological or air quality measurement nodes. For the continuous heart beat menu selection, the Phy Data Acquisition panel is opened, the plethysmography parameters being displayed (Figure 6(c)). Figure 6

Implemented SmartPhone Java2ME panels: (a) find service; (b) select service and (c) physiological parameters service (see online version for colours)

(a)

(b)

(c)

150


Referring to the physiological parameter measurement, two options are available: •

current values display

•

time interval parameters evolution display that permits to visualise the values obtained for an imposed time interval (e.g., 60 s).

When according to imposed event markers stored in the phone’s memory an anomalous situation or a health risk status is detected, a SMS generation is carried out and the SMS is sent upon user permission. Air quality monitoring is the second implemented service. The voltage values from the temperature and relative humidity measuring channels are read continuously or one-shot type according to the user selection. Imposing levels of indoor air conditions, the measured values that are out of the recommended ranges are automatically signalled. Considering that the air quality measuring node delivers the measured values as voltage values, the voltage to physical units conversion (temperature and humidity) is implemented by the smart phone software. The IR and red measuring channels performance evaluation was carried out. By using a modular arbitrary signal generator (NI PXI-5401), a set of simulated photoplethysmography waveforms was applied to the buffer and high-pass filter block and Heart-Rate (HR) and oxyhemoglobin saturation parameter (RSpO2) calculation errors are estimated. For the simulated cases, characterised by 40~250 bpm, the HR relative error was less than 2% from the imposed HR simulated values, while RSpO2 estimation errors were less than 4% of the reading. The beat-to-beat time interval calculation accuracy was obtained using photoplethysmography-type signals generated by a special function generator. The estimation errors of the beat-to-beat time interval values were less than 1% of the imposed values. Additional tests performed on youngsters confirm the values obtained with the simulated signals. Using the physiological parameters measurement nodes, experimental data were obtained from different patients, stored in smart phone memory and downloaded to the laptop PC. The evolution of heart rate and SpO2 are presented in Figure 7 whereas Figure 8 presents the evolution of indoor temperature and humidity during the health status parameters measurement. Figure 7

The evolution of heart rate (HR) and oxygen saturation (SpO2) values during monitoring based on Bluetooth distributed sensing system

Health status and air quality parameters Figure 8

6

151

The evolution of indoor air condition parameters during monitoring based on Bluetooth distributed sensing system (RH – relative humidity, T – temperature)

Conclusion

Computerised tools can be used with EMRs to identify, intervene early in, and track the frequency of adverse health events. A Bluetooth WPAN distributed instrument for health status monitor through plethysmography that can be used in Home TeleCare applications was developed. The work presents an embedded plethysmograph implementation and the developed Java2Me software associated with the measurements and data processing tasks in a distributed system based on a smart phone. Primary data processing is distributed between a microcontroller-based node and a smart phone whereas the advanced processing is performed at a laptop PC level. Events signalling based on SMS alert was also developed. Several tests of the distributed instrument components were carried out proving the capabilities of the implemented solution. Future work is related with the extension of the physiological parameters to be measured, which requires the implementation of new measuring channels and new measuring nodes. In order to extract possible correlations between indoor air quality and HRV new tests will be done under different air conditions and for extended number of volunteers of both sexes characterised by different ages and different respiratory clinical histories.

References Bellazzi, R., Montani, S., Riva, A. and Stefanelli, M. (2001) ‘Web-based elemedicine systems for home-care: technical issues and experiences’, Computers Methods and Programs in Biomedicine, Vol. 64, pp.175–187. Boychuk, R.B., Yamamoto, L.G., Demesa, C.J. and Kiyabu, K.M. (2006) ‘Correlation of initial emergency department pulse oximetry values in asthma severity classes (steps) with the risk of hospitalization’, Am. J. Emerg. Med., Vol. 24, No. 1, pp.48–52. Chan, L.L., Zhang, J., Narayanan, M., Celler, B.G. and Lovell, N.H. (2008) ‘A health monitoring and evaluation system for assessing care needs of residents in aged-care facilities’, Proceedings of BioMed’08, Insbruck, Austria, pp.414–418. Chaudhry, B., Wang, J., Wu, S., Maglione, M., Mojica, W., Roth, E., Morton, S.C. and Shekelle, P.G. (2006) ‘Systematic review: impact of health information technology on quality, efficiency, and costs of medical care’, Ann. Intern. Med., Vol. 144, pp.742–752.

152


de Jode, M. (2005) Symbian on Java, version 2.05, on-line at http://www.symbian.com/files/rx/ file7415.pdf Gao, T., Massey, T., Selavo, L. Bor-rong Chen, Lorincz, K., Shnayder, V., Hauenstein, L., Dabiri, F., Jeng, J., Chanmugam, A., White, D., Sarrafzadeh, M. and Welsh, M. (2007) ‘David Crawford ‘The advanced health and disaster aid network: a light-weight wireless medical system for triage’’, IEEE Transactions on Biomedical Circuits and Systems, Vol. 1, No. 3, pp.203–216. Gao, T., Pesto, C., Selavo, L., Chen, Y., Tia Gao Pesto, C., Selavo, L., Chen, Y., Ko, J.G., Lim, J.H., Terzis, A., Watt, A., Jeng, J., Chen, B-R., Lorincz, K. and Welsh, M. (2008) ‘Wireless medical sensor networks in emergency response: implementation and pilot results’, Proc. IEEE Conference on Technologies for Homeland Security, Aid Networks, Rockville, MD, Vol. I, pp.187–192. Gómez, E.J., Hernando, M.E., García, A., Del Pozo, F., Cermeño, J., Corcoy, R., Brugués, E. and De Leiva, A. (2002) ‘Telemedicine as a tool for intensive management of diabetes: the DIABTel experience’, Computers Methods and Programs in Biomedicine, Vol. 69, pp.163–177. Hillestad, R., Bigelow, J., Bower, A., Girosi, F., Meili, R., Scoville, R. and Taylor, R. (2005) ‘Can electronic medical record systems transform health care? Potential health benefits, savings, and costs’, Health Aff. (Millwood), Vol. 24, pp.1103–1117. Kamat, V. (2002) ‘Pulse oxymetry’, Indian Journal of Anaesthesia, Vol. 46, No. 4, pp.261–268. Khalifa, Y. and Novacky, G. (2006) PDA Programming in C, Kendall/Hunt Publishing Co, August, Dubuque, IA, USA. Konstantas, D., van Halteren, A. and Bults, R. (2004) ‘Mobile patient monitoring: the MobiHealth system’, Stud. Health Technol. Inform, Vol. 103, pp.307–314. Kumar, C., Kline, P. and Thompson, T. (2004) Bluetooth Application Programming with Java APIs, Morgan Kaufmann Publishers, San Francisco, USA. Kuperman, G., Teich, J. and Bates, D.W. (1997) ‘Improving care with computerized alerts and reminders’, Assoc. Health Serv. Res., Vol. 4, pp.224–225. Kyriacou1, P.A., Shafqat1, K. and Pal, S.K. (2007) ‘Arterial blood oxygen saturation during blood pressure cuffinduced hypoperfusion’, Journal of Physics: Conference Series, Vol. 85, pp.12–26. Li, Y-T. (2007) ‘Pulse oximetry’, SEPS Undergraduate Research Journal, Vol. 2, No. 6, on-line at http://www.eps.surrey.ac.uk/surj/ Mendelson, Y. and Ochs, B.D. (1988) ‘Noninvasive pulse oximetry utilizing skin reflectance photoplethysmography’, IEEE Transactions on Biomedical Engineering, Vol. 35, pp.798–805. Newandee, D. and Reisman, S. (2003) ‘Wavelet representation comparison for heart rate variability analysis’, Bioengineering Conference, 2003 IEEE 29th Annual, Proceedings of Publication, March, pp.112–113. Ortiz, E. (2004) Using the Java APIs for Bluetooth Wireless Technology, December, On line at http://developers.sun.com/mobility /apis/articles/bluetoothintro/ Petersen, L.A., Orav, E.J., Teich, J.M., O’Neil, A.C. and Brennan, T.A. (1998) ‘Using a computerized signout program to improve continuity of inpatient care and prevent adverse events’, Jt. Comm. J. Qual. Improv., Vol. 24, pp.77–87. Postolache, O., Girão, P.M., Postolache, G. and Dias Pereira, J.M. (2007) ‘Vital signs monitoring system based on emfi sensors and wavelet analysis’, Proc IEEE Instrumentation and Measurement Technology Conf., Warsaw, Poland, Vol. I, May, pp.3–7. Postolache, O., Postolache, G. and Girão, P.M. (2006) ‘Non-invasive mobile homeostasis instrument’, Proc. IEEE International Workshop on Medical Measurements, Benevento, Italy, Vol. I, May, pp.94–98. Rahman, F., Kumar, A., Nagendra, G. and Gupta, G. (2005) ‘Network approach for physiological parameters measurement’, IEEE Transactions on Instrumentation and Measurement, Vol. 54, No. 1, pp.337–347.


153

Rind, D.M., Safran, C., Phillips, R.S., Wang, Q., Calkins, D.R., Delbanco, T.L., Bleich, H.L. and Slack, W.V. (1994) ‘Effect of computer-based alerts on the treatment and outcomes of hospitalized patients’, Arch. Intern. Med., Vol. 154, pp.1511–1517. Schoen, C., Osborn, R., Trang Huynh, P., Doty, M., Peugh, J. and Zapert, K. (2006) ‘On the front lines of care. Primary care doctors’ office systems, experiences, and views in seven countries’, Health Aff (Mildwood), Vol. 25, pp.w555–w571. Sybase (2008) Healthcare Mobile Solution, On-line at http://www.sybase.co.uk/products/ healthcare/mobilehealthcare Task Force of the European Society of Cardiology and The North American Society of Pacing and Electrophysiology (1996) ‘Heart rate variability. Standards of measurement, physiological interpretation, and clinical use’, European Heart Journal, Vol. 17, pp.354–381. von Kanel, R., Loredo, J.S., Powell, F.L., Adler, K.A. and Dimsdale, J.E. (2005) ‘Short-term isocapnic hypoxia and coagulation activation in patients with sleep apnea’, Clin. Hemorheol. Microcirc., Vol. 33, No. 4, pp.369–377. World Health Organization (2001) International Classification of Functioning, Disability and Health, On-line at http://www.who.int/classifications/icf/site/onlinebrowser/icf.cfm