Career

Influence of Accelerometer Type and Placement on Physical Activity Energy Expenditure Prediction in Manual Wheelchair Users

Description
Influence of Accelerometer Type and Placement on Physical Activity Energy Expenditure Prediction in Manual Wheelchair Users
Categories
Published
of 15
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  RESEARCHARTICLE Influence of Accelerometer Type andPlacement on Physical Activity EnergyExpenditure Prediction in Manual WheelchairUsers TomEdwardNightingale,Jean-PhilippeWalhin,DylanThompson,JamesLeeJohnBilzon * Centrefor DisAbilitySportandHealth(DASH),Departmentfor Health, UniversityofBath,Bath,Somerset,UnitedKingdom *  J.Bilzon@bath.ac.uk Abstract Purpose To assess thevalidityof twoaccelerometer devices, at two different anatomical locations,for the prediction of physical activity energy expenditure (PAEE) in manual wheelchair users (MWUs). Methods SeventeenMWUs (36 ± 10yrs, 72 ± 11kg) completedtenactivities;resting, folding clothes,propulsionona 1% gradient (3,4,5,6and 7 km  hr  -1 ) and propulsionat 4km  hr  -1 (with anaddi-tional 8% bodymass,2% and 3% gradient) ona motorised wheelchair treadmill. GT3X+and GENEActivaccelerometers werewornontheright wrist (W) and upper arm(UA). Line-arregression analysis was conducted between outputs from eachaccelerometer andcrite-rion PAEE, measuredusing indirect calorimetry.Subsequent error statisticswerecalculated for thederived regressionequations for all four device/location combinations,using a leave-one-outcross-validationanalysis. Results Accelerometer outputs at eachanatomical location were significantly (  p < . 01 ) associatedwithPAEE(GT3X+-UA;  r  =0.68andGT3X+-W;  r  =0.82.GENEActiv-UA;  r  =0.87andGEN-EActiv-W;  r   = 0.88).Mean ± SD PAEE estimation errorsfor all activitiescombinedwere 15 ± 45%, 14 ± 50%, 3 ± 25% and 4 ± 26% for GT3X+-UA, GT3X+-W, GENEActiv-UA and GEN-EActiv-W, respectively. Absolute PAEE estimationerrors for devices varied, 19to 66% for GT3X+-UA, 17to 122% for GT3X+-W, 15to 26% for GENEActiv-UA and from 17.0 to 32%for the GENEActiv-W. PLOSONE|DOI:10.1371/journal.pone.0126086 May8,2015 1/15 OPENACCESS Citation:  Nightingale TE, Walhin J-P, Thompson D,Bilzon JLJ (2015) Influence of Accelerometer Typeand Placement on Physical Activity EnergyExpenditure Prediction in Manual Wheelchair Users.PLoS ONE 10(5): e0126086. doi:10.1371/journal.pone.0126086 Academic Editor:  Maciej Buchowski, Vanderbilt University, UNITED STATES Received:  December 8, 2014 Accepted:  March 30, 2015 Published:  May 8, 2015 Copyright:  © 2015 Nightingale et al. This is an openaccess article distributed under the terms of theCreative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in anymedium, provided the srcinal author and source arecredited. Data Availability Statement:  All relevant data arewithin the paper and its Supporting Information files. Funding:  The authors have no support or funding toreport. Competing Interests:  The authors have declaredthat no competing interests exist.  Conclusion Theresults indicate that the GENEActivdevice worn oneither the upper arm or wrist pro-videsthemost valid prediction of PAEE inMWUs. Variation inerror statisticsbetweenthetwodevices isa result of inherent differencesin internal components, on-board filteringpro-cesses andoutputs of eachdevice. Introduction The positive contribution of regular physical activity (PA) to weight balance, metabolic regula-tion and cardiovascular fitness is well documented and broadly accepted in the able-bodiedpopulation [1]. Epidemiological studies concerning the impact of PA on health in wheelchairusers with a spinal cord injury (SCI) have been limited to assessing associations between sub- jective reports of activities of daily living (ADL) [2] or leisure time physical activity (LTPA) [3] and chronic disease risk factors. The assessment of these constructs in previous studies werequantified via the Physical Activity Recall Assessment for people with a SCI [PARA-SCI; [4]],which is administered via a telephone interview. This subjective PA assessment tool asks par-ticipants to recall activities undertaken during the previous 3 days, which is a relatively shortmonitoring period. The results, like other self-report measures, depend on the accuracy of theparticipants ’  memory and recall [5]. Ullrich  et al  , [6] also suggested that the use of the PARA--SCI might have limited application for other investigators besides the developers due to the ex-clusion of subjective appraisals and the technical complexity of interview administration. Todate, despite the aforementioned limitations, quantifying free living PA among wheelchairusers has mostly been restricted to self-report measurements. As such there is a requirement todevelop unobtrusive objective measurement tools that can be easily used to characterise the as-sociation between physical activity and metabolic health in this population.The PA monitoring literature has evolved rapidly, particularly in able-bodied populations,yet there is a paucity of research focussing on the use of activity monitors to predict physical ac-tivity energy expenditure (PAEE) in certain populations, including manual wheelchair users(MWUs). Broadly speaking, various devices used previously to determine PA in MWUs havedistinct limitations. For example despite being unobtrusive, a tri-axial accelerometer attachedto a wheelchair [7] is unable to distinguish between self or assisted propulsion and is also un-able to quantify any activity out of the wheelchair, or during stationary arm crank ergometry exercise. Monitors attached on the waist near the participants ’  centre of mass, as advised by manufacturers for use in able-bodied cohorts, have been shown to under-estimate energy ex-penditure by 24% in MWUs [8]. Previous work by our research group [9] and others [10] has indicated that the anatomical placement location is critical to accurately estimate PAEE. It isperhaps not surprising due to restricted patterns of movement that a tri-axial accelerometerworn on the wrist explains more of the variance in PAEE, resulting in the lowest random errorcompared to the waist. The use of multi-sensor devices has mostly been limited to laboratory-based validation of the SenseWear Armband (SWA). With the development of specific energy expenditure (EE) prediction equations for MWUs [11] the SWA ’ s accuracy has been improvedsomewhat. However, even when using these specific prediction equations, the SWA tended tooverestimate EE (27 to 43%) whereas a wrist-mounted accelerometer accurately predicted EE(9 to 25%) during wheelchair propulsion [12]. Also a recent free-living study using doubly-la-belled water demonstrated that the SWA was unable to detect variation in within-individualEE during voluntary increases in physical activity in individuals with a SCI [13]. PhysicalActivityEnergyExpenditureinManualWheelchairUsersPLOSONE|DOI:10.1371/journal.pone.0126086 May8,2015 2/15  The technological advancements in the field of PAEE assessment have stimulated the devel-opment of sensitive tri-axial accelerometers, capable of storing higher resolution raw, unfilteredacceleration signals, with increased memory capacity for capturing data over prolonged periodsof time [14]. Previous work validating objective PA monitoring tools in MWU ’ s have only re-ported accelerometer outputs as arbitrary count values. The GENEActiv device gives raw accel-eration values, reporting signal vector magnitude (SVM) in  g  -seconds (  g   s -1 ). This device is notsubject to on-board manufacturer-defined band pass filters and hence does not demonstratethe plateau phenomenon of the GT3X+ observed previously [9]. Furthermore, it remains to beestablished whether fluctuations in criterion PAEE during wheelchair propulsion over differing gradients or during load carriage can be detected by accelerometer outputs at the wrist. There-fore, the aim of the present study was to assess the validity of two commonly used accelerome-ter devices, at two different anatomical locations, for the prediction of PAEE in MWUs in acontrolled laboratory environment. Methods Ethics approval was granted by the University of Bath Research Ethics Approval Committeefor Health (REACH) and written informed consent was obtained from each participant. Seven-teen (n = 17) male manual wheelchair users visited the Centre for DisAbility Sport and Health(DASH) human physiology laboratory on one morning following an overnight fast. Ten of theparticipants had complete paraplegia with lesion levels ranging from T1 to L4. Other condi-tions responsible for use of a wheelchair included Spina Bifida (n = 3), Cerebral Palsy (n = 1)and Scoliosis (n = 1). A bilateral lower limb amputee (n = 1), who was considered a regularwheelchair user ( > 70% of locomotion manual wheelchair propulsion) and an able-bodiedwheelchair basketball player (n = 1) who was also familiar with wheelchair propulsion ( > twoyears) were also included in the analysis. Previous work from our research group demonstratedthat the inclusion of numerous disabilities had no measurable impact on the prediction of PAEE in MWUs [9]. Other research has also included participants with various aetiologies re-sponsible for wheelchair use when assessing methods to predict EE in this population [15]. If anything, this approach provides a robust model for the assessment of PAEE in the broaderMWU population and the inclusion of a diverse range of participants is in accordance withbest practice recommendations for PA validation studies [16].Time since injury (TSI) was self-reported based on when the medical condition was first di-agnosed by a clinician. The mass of the wheelchair and participant was recorded in light cloth-ing to the nearest decimal place using platform wheelchair scales (Detecto BRW1000,Missouri, USA). The wheelchair, along with participants shoes were weighed separately andsubtracted from the total mass of the participant plus wheelchair to derive an accurate body mass (kg) [17]. Waist and hip circumference (cm) were measured in duplicate, with partici-pants lying flat on a hard physiotherapy bed, using a metallic tape measure (Lufkin, US). Themean value was taken, and waist/ hip ratio calculated. Skinfold thickness (mm) at 4 upperbody sites (biceps, triceps, subscapular, suprailiac) were measured in duplicate using a set of skinfold calipers (Holtain Ltd, Crymych, UK); the mean value was calculated. Resting metabol-ic rate) (RMR; kcal  day  -1 ) was measured in a semi-recumbent position in accordance with bestpractice [18] using a TrueOne 2400 computerized metabolic system (ParvoMedics, Salt LakeCity, UT). Blood pressure (mmHg) was also measured using an automated blood pressuremonitor (Boso Medicus Prestige, Bosch + Sohn, Germany) in a semi-recumbent positions fol-lowing RMR. Three readings were taken and the mean value was calculated. PhysicalActivityEnergyExpenditureinManualWheelchairUsersPLOSONE|DOI:10.1371/journal.pone.0126086 May8,2015 3/15  Activity Protocol Following the measurement of RMR and anthropometric assessment, participants performedan activity protocol, which consisted of wheelchair propulsion and a folding clothes task, repre-sentative of an activity of daily living. Participants continuously untangled t-shirts placed on adesk, then neatly folded and stacked them. Wheelchair propulsion took place on a wheelchairadapted treadmill with necessary safety features and stabilising arm (HP Cosmos Saturn 250/100r, HaB International Ltd, UK). A counterbalanced approach for randomly assigning theorder of activities was not utilised in this study based on observations from previous work [9].Even with five minutes of recovery in between activities, a considerable carryover effect in oxy-gen uptake (V   ̇ O 2 ) and heart rate (HR) was observed. Therefore, each activity was assigned inorder of intensity and lasted for six minutes interspersed with four minute recovery periods.Wheelchair propulsion on the adapted treadmill included eight trials 3 km  hr -1 , 4 km  hr -1 , 5km  hr -1 , 6 km  hr -1 & 7 km  hr -1 on a 1% gradient. Following a ten minute rest, participants pro-pelled at 4 km  hr -1 on a 1% gradient with 8% of body mass added to the chair in a rucksack and4 km  hr -1 on a 2% and 3% gradient. Instrumentationandassessmentofenergyexpenditure The GT3X+ activity monitor (Actigraph, Pensacola, FL, USA) records time-varying accelera-tions within the dynamic range of ± 6  g  , and contains a solid state tri-axial accelerometer. TheGT3X+ activity monitor is compact (dimensions: 4.6 cm x 3.3 cm x 1.9 cm), lightweight (19grams), and can easily be worn at multiple locations on the body. To quantify the amount andfrequency of human movement, accelerometer outputs are digitized via a twelve-bit analog todigital converter (A/DC) and passed through Actigraph ’ s proprietary digital filtering algo-rithms. In order to eliminate any acceleration noise outside of the normal human activity fre-quency, digitized signals pass through low (0.25 Hz) and high (2.5 Hz) band width filters [19]. ‘ Physical activity counts ’  (PAC) are calculated through summing the change in raw accelera-tion values measured during a specific interval of time, or  ‘ epoch ’ . The GENEActiv tri-axial de- vice (GENEActiv, Activinsights, Cambridge, UK) contains a ± 8  g   seismic accelerometer and isalso compact (dimensions: 4.3 cm x 4.0 cm x 1.3 cm) and lightweight (16 grams). This devicehas been explained in detail elsewhere [20]. Both devices were initialised with a sampling fre-quency of 30 Hz.Throughout the activity protocol two GT3X+ units were worn, one on the right wrist (W,using a Velcro wrist strap positioned over the dorsal aspect of the wrist midway between the ra-dial and ulnar Styloid processes) and one on the right upper arm (UA, using a small elastic beltpositioned on the lateral surface of the arm midway between the acromion process and lateralepicondyle of the humerus). Two GENEActiv accelerometers were also worn; one distal to theGT3X+ on the right W and one on the posterior aspect of the midpoint on the right UA, se-curely fixed to the skin over the triceps brachii using a 10 cm 2 patch of tape (Hyperfix self-ad-hesive dressing retention tape, Smith & Nephew Healthcare Ltd., UK). The GENEActiv andGT3X+ devices were both initialised with a sampling frequency of 30Hz.Continuous gas exchange measurements were collected during each activity, using aTrueOne 2400 computerized metabolic system,calibrated according to manufacturer ’ s instruc-tions prior to use. Fractions of oxygen and carbon dioxide were measured via a paramagneticoxygen analyser and an infrared, single beam, single wave-length carbon dioxide analyser, re-spectively. Metabolic data was retrieved and analysed using associated software (TrueOne met-abolic software, Salt Lake City, UT). V   ̇ O 2  and carbon dioxide production (V   ̇ CO 2 ) were used toestimate EE (kcal . min -1 ) of each activity, using indirect calorimetry. A Polar Team HR monitor(Polar Electro Inc., Lake Success, NY, USA) was also worn during each activity. Accelerometer PhysicalActivityEnergyExpenditureinManualWheelchairUsersPLOSONE|DOI:10.1371/journal.pone.0126086 May8,2015 4/15  outputs and physiological variables for each task and each participant can be found in the sup-porting information file (S1 Dataset). Peak oxygen uptake (V   ̇ O 2  peak) was determined at theend of the activity protocol using a continuous, progressive intensity test with three minute ex-ercise stages until the point of volitional exhaustion. This was conducted using an electrically braked arm crank ergometer (Lode Angio, Groningen, Netherlands) using a prescribed crank rate of 70 rev   min -1 . Statistical Analyses An  a priori  power calculation revealed a sample size of fifteen was necessary in order to detecta correlation coefficient ( r)  of 0.67 using a one-tailed test with an α = 0.05 and power = 0.95.This calculation was based on activity count (counts  s -1 ) from a Computer Science and Appli-cations (CSA) accelerometer and V   ̇ O 2  data during a protocol which included three wheelchairpropulsion velocities [21]. Activity monitors were synchronised prior to use. Breath-by-breathdata was exported into Microsoft Excel from the TrueOne metabolic software and averagedover the final two minutes of each activity. Assuming that dietary-induced thermogenesis wasnegligible (participants came into the laboratory following a 10-hr overnight fast) resting meta-bolic rate (kcal  min -1 ) was subtracted from total energy expenditure (TEE) measured by theTrueOne 2400 computerized metabolic system to generate PAEE for each activity. Compari-sons between the  ‘ criterion ’  measurement of PAEE [TEE — RMR] and activity monitors weremade between the final two minutes of each activity. Data Modelling.  GT3X+ and GENEActiv units were downloaded using ActiLife software(Actigraph, Pensacola, FL, USA) and GENEActiv PC software (version 1.2.1, Activinsights,Cambridge, UK), respectively. Data was exported to Microsoft Excel in a time and datestamped comma-separated value (c.s.v.) file format. Physical activity counts (counts  min -1 )from the GT3X+ and Signal vector magnitude (SVM  g  s ) data from the GENEActiv were sum-mated into 60-s epochs. Activity counts (counts  min -1 ) from the GT3X+ and raw acceleration values (  g   min -1 ) from the GENEActiv were then averaged over the final two minutes of eachactivity. Physical activity energy expenditure prediction models were developed using corre-sponding data from each task for each device at each location, using linear regression analysis.The dependent variable was PAEE (kcal  min -1 ) during the final two minutes of each task (thatis, 171 values in total). The independent variables included accelerometer outputs, i.e.counts  min -1 and SVM  g  . min-1  for the GT3X+ and GENEActiv, respectively. Pearson productmoment correlation coefficients ( r  ) and coefficients of determination (R  2 ) statistics were con-ducted to assess the association between the criterion PAEE and outputs from each device ateach location. Standard Error of the Estimate (SEE) was also calculated for each correlation. Error statistics.  When the development and evaluation of predictive models are con-ducted on the same participants, subsequent evaluation statistics tend to be biased and areoften overly optimistic [22]. Hence, there is a need to cross-validate generated prediction equa-tions using an independent sample; this can be achieved by using a  ‘ split sample ’  approachwhereby half of the participants are used for developing the models and half for cross-valida-tion. However, it is not always feasible to utilise this  ‘ split sample ’  approach when sample sizesare small, a common occurrence when working with disabled populations due to challengeswith participant identification and recruitment [23]. This problem was overcome by develop-ing the estimation algorithm on all but one of the participants and then evaluating it on the ‘ held-out ’  participant by calculating the PAEE prediction error. This process was repeatedwhere each participant acted as the held-out participant and the mean of all the error evalua-tions were calculated. This procedure is a  ‘ leave-one-out cross validation ’  described in more de-tail elsewhere [24]. PhysicalActivityEnergyExpenditureinManualWheelchairUsersPLOSONE|DOI:10.1371/journal.pone.0126086 May8,2015 5/15
Search
Similar documents
View more...
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks