Acquisition of Multiple Physiological Parameters

International Journal of E-Health and Medical Communications, 2(4), 37-49, October-December 2011 37

Copyright © 2011, IGI Global. Copying or distributing in print or electronic forms without written permission of IGI Global is prohibited.

Keywords: Acquisition System, Experimental Method Procedure, Physical Activity, PhysiologicalParameters,Sensors

INtroDuCtIoN

The level of daily physical activity has a strong effect on a person’s quality of life and health. During the last decades, sedentary and inactive lifestyles have raised serious concerns regard-ing public health, with diseases such as cardiac

illnesses, diabetes, obesity and others receiving a strong concern focus, as it is estimated by the World Health Organization (WHO) that obesity alone is responsible for 6% of the health budget of the European nations (WHO, 2006). The industrial development and its employment growth have in general significantly reduced the physical activity of the active population and the technological development has also

Acquisition of Multiple Physiological Parameters During Physical exercise

VirginieFelizardo,UniversityofBeiraInterior,Portugal

PedroDinisGaspar,UniversityofBeiraInterior,Portugal

NunoM.Garcia,LusophoneUniversityofHumanitiesandTechnologies(ULHT),Portugal

VictorReis,UniversityofTrás-os-MontesandAltoDouro,Portugal

AbStrACtThispaperdescribestheexperimentalmethodfocusedontheacquisitionofvariousphysiologicalparametersduringdifferenteffortlevelsofphysicalexerciselikewalkingandrunningatseveralvelocities.Thestudyinvolved57youngandadultpeople,43maleand14female(24,37±5,96years),fromwhich48weresoldiersbelongingtotheInfantryRegimentn.°13(RI13)ofthePortugueseArmyand9wereteachersorcollegestudentsofSportSciences,physicallyactivebutnotcompetitive.Theexperimentalmeasuresprovideasetofinformationthatoffersinsightaboutthehealthstatusandphysicalperformanceofthesubjectsduringexercise.ThisexperimentalmethodprocedureissuitedfortheacquisitionofphysiologicalparameterswithboththewirelessphysiologicaldataacquisitionsystemssuchasthebioPluxandtherespiratoryanalyzergassystemssuchasCosmedK4b2.Thedatawascollectedtoallowthedefinitionofamodelthatwillbeusedtoestimatetheenergyexpenditureofasubjectusingawirelessphysiologicaldataacquisitionsystem,whichismuchmorecomfortableandsuitabletomonitorphysicalexerciseineverydayusethanthestandardmethodthatmakesuseofarespiratorygasanalysissystem.

DOI: 10.4018/jehmc.2011100103

38 International Journal of E-Health and Medical Communications, 2(4), 37-49, October-December 2011


allowed the increased number and increased degree of sedentary life styles (Laggeros & Lagiou, 2007). A significant part of the ac-tive population in developed countries lead a life with little or no physical activity and its leisure time is often occupied with sedentary activities like reading, surfing on the Internet, playing computer games, watching television, among others.

Currently there are a number of devices who assess the physical activity of an individual, for example the ActiGraph Activity Monitor (ActiGraph, 2010) or the RT3 Accelerometer (Stayhealthy, 2010). The goal of these devices is to monitor whether a person is fulfilling his/her amount of daily exercise or not, and to raise alarms if the foreseen levels are not met.

Taken to a next level, the evaluation of the physical activity level of a person has its roots in the need to determine if his/her physical activity behaviour falls within an interval or profile defined by a health care provider, us-ing appropriate personalized criteria, with the objective of maintain or improve his/her health status. It is well known that physical activity reduces the risk of various diseases and brings several benefits, including improved glucose metabolism, reduced body fat and lower blood pressure (Choi etal., 2005).

The goals and expected benefits of physical activity monitoring, underline the importance of the acquisition of physiological parameters during exercise.

However, until very recently, it was difficult to get a set of physiological data using several measuring devices that are typically in different locations, some of them causing discomfort to those who use them, many of these having an associated cost of ownership and or operation that are not compatible to the desirable goal of enlarging its use to a broad sector of the population, thus rendering impracticable its use in Assisted Living solutions.

There are several methods and techniques to estimate the energy expenditure and the physi-cal activity level of a subject. The techniques of direct calorimetry, indirect calorimetry and doubly-labeled water are considered the

most precise methods for measuring the level of metabolism. Direct calorimetry measures the energy expended by assessing the rate of body heat lost to the environment, through either aerobically or anaerobically forms, and is held in chambers. This method has some disadvantages such as its high cost, the need for a long experimental time and the need for an artificial controlled environment, which is impracticable as part of a daily routine. Doubly-labelled water is one of the methods considered as “gold standard” for determining the energy expenditure (Murphy, 2009). It consists in the ingestion of water labelled with deuterium and oxygen isotopes. The measurement of the con-centration of these elements in excreted urine and exhaled air enables the calculation of the energy demand by the organism of that subject. Its high cost (Littlewood etal., 2002) and the need for specialized personnel and equipment, has restricted its use in further studies. This method still has additional disadvantages since it only provides information on the total energy expenditure, and cannot be accurately used to monitor a subject in a small time interval. Moreover, the method gives no information about the type, intensity and duration of physical activity (Yamada etal., 2009). Thus, the defacto “gold standard” for the assessment of physical activity is the analysis of respiratory gases by indirect calorimetry (Choi et al., 2005). The respiratory gases, in particular, the production of gases such as carbon dioxide (CO2), is a good indicator of the amount of metabolic activity going on in the cells, because this gas is a direct result of oxidation processes. Oxygen consump-tion and carbon dioxide release is measured and the Respiratory Exchange Ratio (RER) is calculated to establish which type of food is being oxidized. It is assumed that each food type releases a given amount of energy for each liter of oxygen consumed, and that the body’s oxygen and carbon dioxide contents remain constant. Total energy expenditure is estimated from the RER and from total oxygen consump-tion (Littlewood etal., 2002). The CosmedK4 family of devices performs this analysis. The CosmedK4 device is able to recollect all the



breathing gases from the subject, analyse its contents and report this data to a computer. Yet, as a mean to assess everyday-life activity and despite of being portable, the CosmedK4b2 gas analysis device is not comfortable for the users, limiting the athletic performance and its usage scenarios (for example, it would be impossible to use it in a Judo exercise).

Others methods for estimating the energy expenditure of daily activities are based on data collection of heart rate and movement techniques. They are often used due to its ease of data collection and its low cost (Crouter etal., 2005). Monitoring energy expenditure by measuring heart rate is based on the relation-ship between energy expenditure and heart rate. Although monitors accurately measure heart rate, its accuracy for the measurement of energy expenditure is limited by the relative independence of heart rate change from physi-cal activity. In this method, the energy cost is estimated from the adjustment of individual curves for a variety of activities in laboratory. Among factors that may alter the relation be-tween Oxygen consumption (VO2) response and the amount of exercise are variation in tem-perature and humidity, fatigue, hydration, and even emotional responses. Another limitation of this method is due to the fact that in sedentary subjects, the measured heart rate in nearly 24 hours does not exceed the limits of the heart rate when the subject is resting, making it dif-ficult to distinguish between light and moderate activities. Despite this limitation, heart rate can provide an indication of the intensity, duration and frequency of activity (Reis etal., 2000).

The use of electronic motion sensors is based on the assumption that the movement of body segments reflects the total energy ex-penditure (Reis etal., 2000). These sensors are placed on the wrist or waist of the subject and measure the energy expenditure by recording the body accelerations over time, allowing calcula-tion of oxygen uptake and energy expenditure from estimation equations. Examples of these instruments are pedometers and accelerometers. A pedometer is a motion sensor that saves uni-axial movements of steps in response to body

acceleration on the vertical axis. The distance of the course taken by the subject can be esti-mated by calibrating the equipment to its stride magnitude. Despite presenting low cost, it does not measure sedentary activities, isometric exercises and arm movements, underestimat-ing distances at low speeds and overestimating distances in walking and running fast (Reis etal., 2000). The accelerometers are motion sensors, sensitive to variations in body acceleration in one or three axels and therefore able to directly and objectively measure the frequency, intensity and duration of the movements and relate them to the physical activity performed.

Some authors have shown that there is an increase of caloric expenditure with physical activity, depending on the activity type (such as walking or running) as well as anthropological characteristics (gender, age, weight and height) (Choi etal., 2005). The energy expenditure es-timation measured by accelerometers assumes that a subject’s movement generates body acceleration, theoretically proportional to the force exerted by the muscles responsible for this acceleration and therefore proportional to the energy expended. The monitoring of physical activity using accelerometers avoids many of the limitations of other methods, such as limitations on cost, portability and limitations on requiring specific setups and assistance personnel, and can be very useful to obtain objective information to estimate energy expenditure on a daily basis with little intervention (Yamada etal., 2009).

Among recent methods that use motion to estimate the energy expenditure, applications for smartphones such as Android, Blackberry, iPhone, and others can be highlighted. These applications map the physical activity of outdoor sports using the Global Positioning System (GPS) of the smartphone. Some examples of these applications are CardioTrainer, Endo-mondo, SportyPal, MyTracks and RunKeeper. Forexample,CardioTrainer (WorkSmart Labs, 2011) is a free application for Androidsmart-phones that tracks and records all the physical activity. This application works in conjunction with GoogleMaps. It takes the route indicated on a map to give information about the distance



travelled in km. It also provides information of average speed, number of calories expended, and travel time. The data is uploaded immedi-ately to the company’s servers. Although this type of application has very low accuracy it has wide applicability because there is no need of hardware (besides the smartphone) and it has demonstrated adherence and motivation from some audience.

Experimental procedures that are more objective and accurate have been published for the above mentioned methods (Oliveira & Maia, 2001), but these also require sophisti-cated and expensive equipment in addition to the rather complex processes of data analysis. Due to its complexity, most of these methods cannot be applied to epidemiological studies, but instead serve as a validation criterion for ground methods.

The questionnaires of physical activity represent one of the easiest methods, practical and economic to estimate physical activity, especially in epidemiological purposes for large samples (Oliveira & Maia, 2001). Although the questionnaires have advantages of lower cost and wider scope of studies when compared to other methods, their use has some disadvan-tages in the survey procedure because they not provide energy expenditure estimation as precisely as other methods and are, of course, more prone to subject induced intentional or unintentional error.

In recent years, considerable progress has been made in communication protocols, sensing and transmission devices, sensor min-iaturization, battery life span, human-machine interfaces and acquisition systems that allow the design of more ergonomical devices and thus its widespread use in everyday life.

Although not focusing on respiratory pro-cesses, this work focus on the amount of energy spent on and by the subject’s movement and it may be used to assess its energy expenditure, in particular when submitted to a given activity protocol. Furthermore, the joint use of simple sensors such as accelerometers and complex gas analysis performed by the Cosmed K4, allows the establishment a comparison of the

data and the creation of a relation between these two measurements.

This paper describes the experimental method of a research study named “Study of validation and collection of biological signals in young and healthy adults” which focused on the acquisition of various physiological parameters during walking and running activities of a group of soldiers belonging to the Infantry Regiment (RI13) of the Portuguese Army. The research tasks described here are part of a wider research described in Felizardo’s Master of Science dis-sertation in Electrotechnical Engineering at the University of Beira Interior, Portugal, entitled ”ValidaçãodoacelerómetroxyzPluxparaes-timaçãodoGastoEnergéticocomaquisiçãodediversosparâmetrosfisiológicos” (in Por-tuguese), related to the validation of xyzPlux accelerometer to estimate energy expenditure with the acquisition of several physiological parameters.

On another perspective, there is a shortage of real biosignals data publicly available for research. This gap is now reduced as the col-lected data is now published and accessible in the web site belonging to the Assisted Living Computing and Telecommunications Labora-tory – ALLab, a research laboratory within the Networks Group of the Institute of Telecom-munications at the University of Beira Interior, Covilhã, Portugal.

This paper is organised as follows: this paragraph ends the research topic introduction and its purpose is to describe the framework and motivation of the developed work; We describe the relevance of this study, present-ing the sensors used, its underlying rationale, and how these measurements contribute to the purpose of the work; we present the data col-lection methodology, including data variables and collection procedures; finally, section we present our conclusion.

reLevANCe of StuDy

The human body comprehends a great variety of processes, some relatively simple, but others



of great complexity. Thus, the simultaneous acquisition of multiple signals and physiological data during physical effort has great relevance. It can provide a set of information on physical performance during exercise and allows it use for further diagnosis of a particular subject’s health status. There is a lack of physiological data available for the research community in general. This experimental component also contributes to reduce this shortage, by making available the collected data for further studies. The data files, conveniently unidentified and dimensionless, are available in a datawiki http://allab.it.ubi.pt/mediawiki. The format of the data files varies with the bioPlux (.txt format) and CosmedK4b2 (.xls files) file types.

The main objective of this research is to create a method that allows non-invasive measurements with personal or professional purposes designed to cover a wide range of applications such as: weight control, promo-tion of physical activity, lifestyle assessment, training and research.

The data obtained with this experimental work can address topics such as (1) monitoring the health condition during exercise using vari-ables such as Heart Rate, electric potentials gen-erated by the heart’s electrical activity through Electro Cardiography (ECG) and others, (2) assessing the physical performance by using variables such as oxygen consumption, energy expenditure, acceleration of movement, bioelec-tric muscle activity through electromyography (EMG) and (3) still emotional support during exercise using the information inferred by the activity of the sympathetic nervous system by measurement of the electro-dermal activity.

The Cosmed K4b2 estimates energy ex-penditure based on the volume of oxygen con-sumed. It provides an estimation of metabolic energy expenditure, which is a physiological factor. However, the accelerometer and EMG signals can provide more information about the additional energy expended in physical activity, which is the largest component of variation depending on each individual (as-sociated with biomechanical factors). The relationship between human movement and

its power consumption depends on the subject characteristics, such as origin and insertion of muscle, muscle length and size, bone structure, flexibility, body mass and even personality. Due to individual characteristics, each person has a way to minimize energy expenditure.

The relevance of the population type, consisting in healthy and young adults, is related to the study objectives, as the study is designed to assess the energy expenditure dur-ing physical exercise (walking and running), and this age group and population profile is among the groups that practice exercise for leisure or professional means. The sample is characterized by a group of military, college students and college teachers of Sport Science bachelor course, resulting in a sample quite homogeneous in the sense that all participants do daily exercise regularly. Considering differ-ent daily exercise levels and different types of activity, it is possible to cover other population that shows more leisure activity than regular physical activity.

MethoDoLogy

The experimental research study “Validation study and acquisition of biological signals in young and healthy adults” consisted of several measurements during physical exercise in the sports physiology laboratory of the Department of Sports Science, Exercise and Health Univer-sity of Tras-os-Montes and Alto Douro (UTAD), Portugal, with controlled both temperature (16 ºC to 22 ºC) and relative humidity (50%). This study is transversal, and the data was collected during a fixed time exercise plan.

experimental Sample

The study involved 57 young and adult people, 43 male and 14 female aged between 18 and 44 years, with a mean age of 24,37±5,96 years. In this population 48 subjects were soldiers belonging to the Infantry Regiment n°13 (RI13) and 9 subjects were college teachers or college students of the Sport Sciences BSc course, all of them physically active but not competitive.



Anthropometric data relating to age, height, weight and fat percentage of participants in the study of both sexes was collected and are detailed in Table 1.

Study variables

The study consisted on the data collection of control, independent and dependent variables. All variables of the study are shown in Table 2.

Instrumentation

The experimental installation consists of a wireless physiological data acquisition system bioPlux (Plux, 2008), which is connected to a triaxial accelerometer, xyzPlux (measuring range ± 3g); an ECG sensor (ECG triodes), ecgPlux (Gain: 1000, band-pass: 0.05 Hz - 30 Hz; common-mode rejection ratio (CMRR): 110 dB); two sensors of electro-dermal activ-ity, edaPlux (low-pass: 3 Hz); a peripheral temperature sensor, tempPlux (0 °C - 100 °C ± 1.5%); and an electromyography sensor, emgPlux (Gain: 1000, band-pass: 25 Hz - 500 Hz, CMRR: 110 dB).

The bioPlux is the device that collects and digitizes the signals from the sensors, transmit-ting them via Bluetooth® (Connectivity - 10m range, with standard adapter; >100m with high range adapter) to a computer, where they are viewed in real time. The channels of the bioPlux have 12 bits, and its sampling frequency is 1 kHz. The bioPlux also has a digital port, a terminal for connecting the AC adapter and charge the internal battery (allows a lifespan of approxi-mately 12 hours), and a channel to connect the reference electrode, which is essential for the proper monitoring of electromyography signal.

The respiratory analyzer gas, CosmedK4b2 system (Cosmed Corporation, 1998) was used to validate the experimental results and to obtain additional variables. The equipment chosen for the practical exercise was the treadmill.

The choice of the equipment is very depen-dent on the study type and goals. In this study, the experimental protocol is focused on physi-cal exercise. Thus, the choice of the exercise equipment and of the acquisition equipment has a great importance for dependent variables.

During the experimental protocol each subject performed an exercise series of walk-

Table1.Anthropometricdataofstudyparticipants

Subject Age(years) Height(cm) Weight(kg) %fat

General 24,37±5,96 171,26±9,38 69,88±12,10 16,21±4,59

Male 24,21±6,24 174,73±7,73 73,22±11,08 14,22±3,41

Female 24,86±5,19 160,60±4,87 59,62±9,18 22,33±4,91

Table2.Control,Independentanddependentvariablesofstudy

ControlvariablesTemperature (16°C to 22°C) and relative humidity (50%)

Ration

Independentvariables Experimental Protocol (Exercise Plan)

Dependentvariables

Oxygen Consumption (VO2) Heart Rate Energy Expenditure

Electric potentials generated by heart’s electrical activity through

ECGSkin temperature Electro-Dermal Activ-

ity (EDA)

Bioelectric muscle activity through electromyography (EMG)

Oscillations of the movement (acceleration amplitude) through Accelerometer



ing and running. The equipment chosen for the exercise practice is the treadmill. Using this equipment to support the physical activity and using accelerometers to measure data from the generated motion, good correlations for energy expenditure related with physical exercise have been obtained (Jacobi etal., 2007).

For the measurement of physiological data during the exercise protocol, it was cho-sen the bioPlux acquisition system because it allows eight acquisition channels, suited for the acquisition of multiple physiological and environmental parameters. Other main features of this device are also relevant for this experi-ment: this device weights 80 g and includes a Bluetooth communication interface thus allowing a great comfort and independence of movements during the exercise.

The CosmedK4b2 was chosen because it is a “gold standard” in the estimation of energy expenditure and despite the discomfort of who uses it, this equipment is portable.

Data Collection Procedure

Data were collected during 14 days from the May the 3rd to the 24th (2010) in the laboratory of Physiology, Department of Sports Science, Exercise and Health, at the Trás-os-Montes and Alto-Douro University, in Vila Real, Portugal.

In each collection day, the subjects were informed about the objectives, the test procedure for the study and the low risk associated with stress test, being the incidence of problems in the order of 0.01% of tests. After this clarifi-cation, the participants signed a consent and willingness form to participate in the study. The subjects completed an International Physical Activity Questionnaire (IPAQ, 2002) of the previous 7 days.

In order to characterize the sample, the height [cm] and body mass [kg] were measured in standing position during a deep breath through a stadiometer attached to an electronic scale Seca 220 (Seca, 2006). Three skinfolds were also measured to estimate the fat percentage. This procedure was done with skinfold calipers for male subjects at the chest, abdomen and

thigh, while for female subjects at the triceps, supra-iliac and thigh.

The acquisition system, bioPlux (Figure 1), was secured to the left side of the user’s waist. The sensors connected to it were arranged, that is, according to the ECG lead placement pro-tocol, the three ECG leads were placed in the horizontal plane precordial position (V3, V4, V5); peripheral body temperature was measured by placing the temperature sensor in the axillary region on the left side; triaxial accelerometry was collected by placing the accelerometer sensor in the supra-inguinal region on the right side; the two electro dermal activity (EDA) sensors were placed on the abdomen and on the left hand (index and middle fingers); and the electro myography (EMG) sensor in the right rectus femoral (anterior thigh muscle).

The choice for sensors placement is re-lated to the type of the study and to the aimed results.

For whole-body movements, the accel-erometer must be placed near the centre of mass to estimate energy expenditure (Godfrey etal., 2008). The waist and hip are the most used locations for placing the accelerometer (Murphy, 2009).

Rahnama etal. (2006) found a progressive increase in EMG amplitude value for muscles in the rectum femoral, biceps femoral, tibialis anterior and lateral gastrocnemius associated with the increased speed of locomotion on the treadmill from 6 to 21 km/h. In this particular experiment, the rectum femoral was chosen since it has the function of hip flexion and leg extension.

For the location of EDA sensors, one was placed in the index and middle fingers of left hand, since the signal of EDA in this area is strong, yet characterized by emotional states. Therefore, the second EDA sensor was located in the abdomen to compare the type of response and because it causes less discomfort in this location.

The temperature sensor was placed in the auxiliary region because it is one of the areas that where exchange takes place during exercise.



Figure 2 shows a subject in the walking phase of the protocol. The bioPlux device can be seen in left side of the subject’s waist; also visible is the breathing mask for the respiratory gas analysis machine.

Exercise Plan

The exercise plan consists in incremental veloci-ties as proposed by Bouten etal. (1994), Choi etal. (2005), and Kim etal. (2009). For men,

Figure1.WirelessphysiologicaldataacquisitionsystembioPluxandsensors(imagecourtesyofhttp://www.plux.info/,notshowninscale)



walking velocity was defined as 5.8 km/h and for women it was set to 5.1 km/h. The three running velocities levels for men were 8.4 km/h, 10.3 km/h, 11.6 km/h and for women were 7.7 km/h, 9 km/h, 10.3 km/h. The duration of each exercise step was five minutes, interspersed with a period of 1 to 3 minutes for recovery. The five minutes exercise duration was defined because the minimum time limit that an exercise must be performed to reach a steady metabolic state is between two to three minutes, depending on the fitness of the individual (Chatagnon & Busso, 2006). This exercise plan includes four different marching / running velocities to allow

the collection of data from physical exercise activities with different intensities. Table 3 shows the velocities of the exercise plan and the correspondent metabolic equivalent (MET) for the classification of activity intensity. Light physical activities vary between 1,5 and 2,99 MET, moderate activities between 3 and 5,99 MET, vigorous activities between 6 and 8,99 MET and very vigorous activities higher than 9 MET (Reis etal., 2000). Differentiated run-ning speeds for different genders were planned as seen before, to allow for different gender metabolic and physical performance.

Figure2.Implementationoftheexerciseprotocolonthetreadmill

Table3.Velocitiesofplanexerciseandthecorrespondentmetabolicequivalent(MET)

Gender Velocity MET Activityclassification

Male 5,8 km/h 5,3±0,9 Moderate

8,4 km/h 8,7±1,6 Vigorous

10,3 km/h 10,1±1,8 Very vigorous


Female 5,1 km/h 4,1±1,0 Moderate

7,7 km/h 7,3±1,7 Vigorous

9 km/h 8,4±1,8 Vigorous




Data

The data obtained from the bioPlux is recorded on open ASCII format (.txt) and from the CosmedK4b2 data is recorded in datasheet files (.xls). In this experimental work were obtained 57 files containing data on each velocity level and the respective pauses.

Each file from bioPlux consists in 10 col-umns. Table 4 shows the description of each column of data.

The data is available from the datawiki from the Assisted Living Computing and Tele-communications Laboratory website available at the http://allab.it.ubi.pt/mediawiki/ address.

CoNCLuSIoN

The ultimate goal of this research was the use of easily collected physiological data to assess the energy expenditure of a healthy subject. The validation of the estimated model was achieved by comparison with the considered “gold standard” of respiration gases analysis. Yet, the tasks that were involved in the obten-tion of the data are per se worthy of description and may serve as guidelines to future research in this area.

To the authors best knowledge this was the first time that such a collection of physiological

parameters has been recorded, allowing com-parison with “gold standard” data, and further allowing research by selection of different aspects of the same exercise protocol.

This research showed that the used method-ology is adequate for the collection and record-ing of the number and variety of physiological parameters. The fact that this methodology has tapped the potential of bioPlux for the acquisi-tion of multiple physiological parameters and the collaboration of the military subjects con-tributed to add value to the recorded data, thus enabling and valuing the availability of data for those who want to further research on it.

Also, the large number of variables ob-tained during exercise allows carrying out work in areas such as monitoring the health condition, athletic performance, and energy expenditure analysis across gender, among other studies.

However, the methodology has limitations related to the experimental sample and exercise plan. The experimental sample can limit the study scope because it is composed by youth and adults with habits of daily physical activ-ity, which is not representative of the general population. Nevertheless, considering an active and homogeneous sample has relevance in this type of study. Furthermore, the plan does not include exercise at slow speeds and levels of sedentary activity as the physical activity per-formed was classified as moderate, vigorous

Table4.ContentsofbioPluxfiles

Column Data

1 Sequential number (repeating from 0 to 127)

2 (not used)

3 x axes of accelerometer (vertical axes)

4 y axes of accelerometer (medial-lateral axes)

5 z axes of accelerometer (anterior-posterior axes)

6 EMG

7 ECG

8 EDA (hand)

9 EDA (abdomen)

10 Skin temperature



and very vigorous. So the further research is needed to collect and record data for light and sedentary activities.

As two acquisition devices were used, being one of them a “gold standard” for the estimation of energy expenditure, the CosmedK4b2, it was possible to validate the xyzPlux accelerometer for the energy expenditure estimation. More-over, the acquisition system bioPlux is shown to be applicable for a wide number of scenarios, in particular in studies involving physical activity or free living, due to its portability, lightweight and wireless communication capabilities.

ACkNowLeDgMeNtS

The authors thank our colleagues in the Depart-ment of Sports Science, Exercise and Health University of Trás-os-Montes and Alto Douro (UTAD), Portugal, Romeu Mendes and Fer-nando Policarpo for their support, availability, technical support and materials provided dur-ing the collecting data. The authors wish to acknowledged the collaboration provided by the constituents of the sample, members of the Infantry Regiment n°13 (RI13). Finally the authors thank Plux Lda. for making available the equipment bioPlux used in this research.

refereNCeS

ActiGraph. (2010). Technicalinformation-GT3X-plusactivitymonitor. Retrieved from http://www.theactigraph.com/

Bouten, C. V., Westerterp, K. R., Verduin, M., & Jans-sen, J. D. (1994). Assessment of energy expenditure for physical activity using a triaxial accelerometer. MedicineandScienceinSportsandExercise, 26(12), 1516–1523. doi:10.1249/00005768-199412000-00016

Chatagnon, M., & Busso, T. (2006). Modelling of aerobic and anaerobic energy production during exhaustive exercise on a cycle ergometer. Euro-pean Journal of Applied Physiology, 755–760. doi:10.1007/s00421-006-0236-3

Choi, J. H., Lee, J., Hwang, H. T., Kim, J. P., Park, J. C., & Shin, K. (2005). Estimation of activity energy expenditure: Accelerometer approach. In Proceedings of the IEEE27thAnnualConferenceonEngineering inMedicineandBiology (Vol. 4, pp. 3830-3833).

Crouter, S. E., Clowers, K. G., & Bassett, D. R. (2005). A novel method for using accelerometer data to predict energy expenditure. EuropeanJournalofAppliedPhysiology, 100, 1324–1331. doi:10.1152/japplphysiol.00818.2005

Felizardo, V. (2010). Validação do acelerómetroxyzPluxparaestimaçãodoGastoEnergéticocomaquisiçãodediversosparâmetrosfisiológicos. Un-published doctoral dissertation, University of Beira Interior, Covilha, Portugal.

Godfrey, A., Conway, R., & Meagher, D., & ÓLaighin, G. (2008). Direct measurement of human movement by accelerometry. MedicalEngineering&Physics, 30, 1364–1386. doi:10.1016/j.medeng-phy.2008.09.005

Guedes, D. P., Lopes, C. C., & Guedes, J. E. R. P. (2005). Reprodutibilidade e validade do questionário internacional de atividade física em adolescentes. RevistaBrasileiradeMedicinadoEsporte, 2(11), 151–158.

International Physical Activity Questionnaire (IPAQ). (2002). Downloads. Retrieved from http://www.ipaq.ki.se/downloads.htm

Jacobi, D., Perrin, A., Grosman, N., Dore, M., Nor-man, S., Oppert, J., & Simon, C. (2007). Physical activity-related energy expenditure with the RT3 and TriTrac accelerometers in overweight adults. Obesity(Silver Spring, Md.), 15, 950–956. doi:10.1038/oby.2007.605

Kim, D. H., Cho, J. S., Oh, H. S., Chee, Y. J., & Kim, I. Y. (2009). The estimation method of physical activity energy expenditure considering heart rate variability. ComputersinCardiology, 36, 413–416.

Lagerros, Y. T., & Lagiou, P. (2007). Assessment of physical activity and energy expenditure in epide-miological research of chronic diseases. EuropeanJournalofEpidemiology, 22, 353–362. doi:10.1007/s10654-007-9154-x

Littlewood, R. A., White, M. S., Bell, K. L., Davies, P. S. W., Cleghorn, G. J., & Grote, R. (2002). Compari-son of the Cosmed K4b2 and the Deltatrac IITM meta-bolic cart in measuring resting energy expenditure in adults. ClinicalNutrition(Edinburgh,Lothian), 2(16), 491–497. doi:10.1054/clnu.2002.0580



Murphy, S. L. (2009). Review of physical activity measurement using accelerometers in older adults: Considerations for research design and conduct. PreventiveMedicine, 48, 108–114. doi:10.1016/j.ypmed.2008.12.001

Oliveira, M. M., & Maia, J. A. (2001). Avaliação da actividade física em contextos epidemiológicos - Uma revisão da validade e fiabilidade do acelerómetro Tritrac–R3D, do pedómetro Yamax Digi-Walker e do questionário de Baecke. RevistaPortuguesadeCiênciasdoDesporto, 1(3), 73–88.

Reis, R. S., Petroski, E. L., & Lopes, A. (2000). Medidasdaatividadefísica:Revisãodemétodos. Flo-rianópolis, Brazil: Revista Brasileira de Cineantro-pometria de Desempenho Humano.

Stayhealthy. (2010). RT3tri-axialresearchactivitymonitor. Retrieved from http://www.stayhealthy.com/

WHO. (2006, November 15-17). Dietandphysicalactivity for health. Paper presented at the WHO European Ministerial Conference on Counteracting Obesity, Istanbul, Turkey.

WorkSmart Labs. (2011). CardioTrainer. Retrieved from http://www.worksmartlabs.com/

Yamada, Y., Yokoyama, K., Noriyasu, R., Osaki, T., Adachi, T., & Itoi, A. (2009). Light-intensity activities are important for estimating physical activity energy expenditure using uniaxial and triaxial accelerom-eters. EuropeanJournalofAppliedPhysiology, 105, 141–152. doi:10.1007/s00421-008-0883-7

VirginiedosSantosFelizardoiscurrentlyaM.Sc.ResearcherattheDepartmentofElectrome-chanicalEngineeringatUniversityofBeiraInterior,Covilhã,Portugal.ShegraduatedinBio-medicalSciences(UniversityofBeiraInterior,Covilhã,Portugal)and,in2010,receivedM.Sc.inElectrotechnicalEngineering–BionicSystems(UniversityofBeiraInterior,Covilhã,Portugal).SheisinvolvedinresearcheractivitiesinAssistedLivingComputingandTelecommunicationsLaboratory–ALLab,aresearchlaboratorywithintheNetworksGroupoftheInstituteofTele-communicationsattheUniversityofBeiraInterior,Covilhã,Portugal.Herresearchinterestsincludesensorsforbiomedicalapplications,biomedicalinstrumentation,Healthmonitoring,Medicalsignalacquisition,analysis,andprocessing.

PedroDinisGasparisanassistantprofessorattheDepartmentofElectromechanicalEngineeringatUniversityofBeiraInterior,Covilhã,Portugal.HegraduatedinElectromechanicalEngineer-ingandreceivedaMScdegreeinEngineeringofEnergyProductionandConservationSystems.In2008hereceivedhisPhD.HehasbeeninvolvedinteachingactivitiesofsubjectslikeControl(DiscreteandDigital)Systems;InstrumentationandMeasurements;DataAcquisition;Indus-trialInformatics;AssistedTherapeuticsandMonitoring;Robotics;EnergyandSustainabilityandRenewableEnergiesinundergraduateandgraduatecourses.Hisresearchinavarietyoftechnologyareas(includingenergyharvestingsystems,energyestimation,assistedtherapeuticsandmonitoringportabledevices,lowpowerembeddedsystemsandrefrigeration)waspublishedinanumberofjournals,conferencepublications,andbooks.HehasparticipatedinseveralNationalandEuropeanresearchprojects.HisresearchinterestsincludeThermodynamicsandHeatTransfer,EnergeticSystems(renewable,production,rationalizationandsustainability),Automation,Robotics,EmbeddedSystemsandControl.



NunoM.GarciaisaProfessorattheSchoolofCommunication,ArtsandInformationTech-nologiesoftheLusophoneUniversityofHumanitiesandTechnologiesinLisbon,Portugal,andInvitedProfessorattheUniversityofBeiraInterior,Covilhã,Portugal.HeholdsaPhDdegreefrom theUniversity ofBeira Interior,Covilhã,Portugal.He is currently coordinator of theAssistedLivingComputingandTelecommunicationsLaboratory,aresearchgroupwithintheInstituteofTelecommunicationspoleattheUniversityofBeiraInterior.HeisalsoaresearcherinCriavisionLda.HismaininterestsincludeNext-GenerationNetworks,advancedalgorithmsforbio-signalprocessing,distributedandcooperativeprotocols.Heismainauthorofseveralinternationalandeuropeanpatents.HeisanISOCmemberandaIEEEMember.VictorMachadoReislecturesExercisePhysiologyattheSportSciences,Exercise&HealthDepartmentoftheUniversityofTrás-os-MontesandAltoDouro(UTAD),PortugalandperformsresearchintheResearchCenterforSports,Health&HumanDevelopment(CIDESD).HeholdsaPhDdegreeonExercisePhysiologyandhismajorinterestsofresearcharethephysiologicalandbiomechani-calindicatorsofenergycostduringphysicalactivities.

VictorReislecturesExercisePhysiologyattheSportSciences,Exercise&HealthDepartmentoftheUniversityofTrás-os-MontesandAltoDouro(UTAD),PortugalandperformsresearchintheResearchCenterforSports,Health&HumanDevelopment(CIDESD).HeholdsaPhDdegreeonExercisePhysiologyandhismajorinterestsofresearcharethephysiologicalandbiomechanicalindicatorsofenergycostduringphysicalactivities.

Acquisition of Multiple Physiological Parameters

Documents

assisted living

infantry regiment

hearts electrical

total energy

daily physical

estimate energy

physical activity

multiple physiological