The relationship between trunk muscle electromyography and lifting moments in the sagittal and frontal planes

In this study, we explore the relationship between moments in the frontal and sagittal planes, generated by a lifting task, vs the electromyographic (EMG) activity of right and left trunk muscle groups. In particular, we postulate that the functional dependence between erector spinae muscle activity and the applied lifting moments about the spine is as follows: the sum of left and right erector spinae processed EMG depends on the sagittal plane moment, and the difference of left and right erector spinae processed EMG depends on the frontal plane moment. A simple out-of-sagittal plane physical model, treating the lumbar spine as a two degree-of-freedom pivot point is discussed to justify these hypotheses. To validate this model, we collected surface EMG and lifting moment data for ten males performing a grid of frontal and sagittal plane lifting tasks. A digital RMS-to-DC algorithm was developed for processing raw EMG. For these tests, we measured EMG for the left and right erector spinae and for the left and right external oblique muscles. The processed EMG signals of the left and right erector spinae muscles are summed and differenced for comparison to the measured sagittal and frontal plane moments. A linear correlation (r2) of 0.96 was obtained for the sum of erector spinae EMG vs the sagittal plane moment; a corresponding value of r2 = 0.95 was obtained for the difference vs the frontal plane moment. No correlations (r2 less than 0.004) was found for the sagittal plane moment and the difference of the left and right erector spinae EMG, and the frontal plane moment and the sum of the left and right erector spinae EMG.     

Development of a novel MATLAB-based framework for implementing mechanical joint stability constraints within OpenSim musculoskeletal models

The Static Optimization (SO) solver in OpenSim estimates muscle activations and forces that only equilibrate applied moments. In this study, SO was enhanced through an open-access MATLAB interface, where calculated muscle activations can additionally satisfy crucial mechanical stability requirements. This Stability-Constrained SO (SCSO) is applicable to many OpenSim models and can potentially produce more biofidelic results than SO alone, especially when antagonistic muscle co-contraction is required to stabilize body joints. This hypothesis was tested using existing models and experimental data in the literature. Muscle activations were calculated by SO and SCSO for a spine model during two series of static trials (i.e. simulation 1 and 2), and also for a lower limb model (supplementary material 2). In simulation 1, symmetric and asymmetric flexion postures were compared, while in simulation 2, various external load heights were compared, where increases in load height did not change the external lumbar flexion moment, but necessitated higher EMG activations. During the tasks in simulation 1, the predicted muscle activations by SCSO demonstrated less average deviation from the EMG data (6.8% −7.5%) compared to those from SO (10.2%). In simulation 2, SO predicts constant muscle activations and forces, while SCSO predicts increases in the average activations of back and abdominal muscles that better match experimental data. Although the SCSO results are sensitive to some parameters (e.g. musculotendon stiffness), when considering the strategy of the central nervous system in distributing muscle forces and in activating antagonistic muscles, the assigned activations by SCSO are more biofidelic than SO.     

Longer static flexion duration elicits a neuromuscular disorder in the lumbar spine.

The objective of this study was to assess the impact of two sequential long, static, anterior lumbar flexions on the development of a neuromuscular disorder and to compare it with previously obtained data from a series of short static flexion periods of the same cumulative time (Sbriccoli P, Solomonow M, Zhou BH, Baratta RV, Lu Y, Zhu MP, and Burger EL, Muscle Nerve 29: 300-308, 2004). Static flexions with loads of 20, 40, and 60 N were applied to the lumbar spine over two 30-min periods with a 10-min rest in between. The reflex EMG activity from the multifidus muscles and supraspinous ligament displacement (creep) was recorded during the flexion periods. Creep and EMG were also monitored over 7 h of rest following the work-rest-work cycle. It was found that the creep that developed in the first 30-min flexion period did not recover completely during the following 10 min of rest, giving rise to a large cumulative creep at the end of the work-rest-work session. Spasms were frequently seen within the EMG during the static flexion. Initial and delayed hyperexcitabilities were observed in all of the preparations at any of the three loads explored during the 7-h rest period. ANOVA revealed a significant effect of time (P < 0.0001) on the postloading data. Larger loads elicited larger magnitudes of the initial and delayed hyperexcitabilities, yet were not statistically different. It was concluded that the 3:1 work-to-rest duration ratio resulted in a neuromuscular disorder, regardless of the load magnitude. The conclusions are reinforced in view of the results from a previous study using 60 min of flexion overall but at 1:1 work-to-rest ratio in which only the highest load elicited a delayed hyperexcitability (Sbriccoli et al., Muscle Nerve 29: 300-308, 2004). An optimal dose-to-duration ratio needs to be established to limit, attenuate, or prevent the adverse effects of static load on the lumbar spine while considering the loading duration as a major risk factor.     

Hybrid neuromusculoskeletal modeling to best track joint moments using a balance between muscle excitations derived from electromyograms and optimization

Current electromyography (EMG)-driven musculoskeletal models are used to estimate joint moments measured from an individual?s extremities during dynamic movement with varying levels of accuracy. The main benefit is the underlying musculoskeletal dynamics is simulated as a function of realistic, subject-specific, neural-excitation patterns provided by the EMG data. The main disadvantage is surface EMG cannot provide information on deeply located muscles. Furthermore, EMG data may be affected by cross-talk, recording and post-processing artifacts that could adversely influence the EMG?s information content. This limits the EMG-driven model?s ability to calculate the multi-muscle dynamics and the resulting joint moments about multiple degrees of freedom. We present a hybrid neuromusculoskeletal model that combines calibration, subject-specificity, EMG-driven and static optimization methods together. In this, the joint moment tracking errors are minimized by balancing the information content extracted from the experimental EMG data and from that generated by a static optimization method. Using movement data from five healthy male subjects during walking and running we explored the hybrid model?s best configuration to minimally adjust recorded EMGs and predict missing EMGs while attaining the best tracking of joint moments. Minimally adjusted and predicted excitations substantially improved the experimental joint moment tracking accuracy than current EMG-driven models. The ability of the hybrid model to predict missing muscle EMGs was also examined. The proposed hybrid model enables muscle-driven simulations of human movement while enforcing physiological constraints on muscle excitation patterns. This might have important implications for studying pathological movement for which EMG recordings are limited.     

Porcine models in spinal research: calibration and comparative finite element analysis of various configurations during flexion-extension

This study was conducted to develop and calibrate a detailed 3-dimensional finite element model of the porcine lumbar spine and to compare this model with various configurations in flexion and extension. Computed tomography scans obtained from the L4-L5 lumbar segment of a Landrace x Large White pig were used to generate a solid volume. The various passive components were characterized by using a step-by-step calibration procedure in which the material properties of the anatomic structures were modified to match the corresponding in vitro data set-points retrieved from the literature. The range of motion of the totally assembled intact model was assessed under a 10-Nm flexion-extension moment and compared with data from a bilateral complete and hemifacetectomy configuration. In addition, the results from our porcine model were compared with published data regarding range of motion in a human finite element model in order to predict the configuration of the porcine model that most closely represented the human spine. Both the intact and hemifacetectomy configurations of the porcine model were comparable to the human spine. However, qualitative analysis of the instantaneous axis of rotation revealed a dissimilarity between the intact porcine model and human spine behavior, indicating the hemifacetectomy configuration of the porcine model as the most appropriate for spinal instrumentation studies. The present 3-dimensional finite element porcine model offers an additional tool to improve understanding of the biomechanics of the porcine spine and to decrease the expense of spinal research.     

Calibration of EMG to force for knee muscles is applicable with submaximal voluntary contractions.

PURPOSE: In this study, the influence of using submaximal isokinetic contractions about the knee compared to maximal voluntary contractions as input to obtain the calibration of an EMG-force model for knee muscles is investigated. METHODS: Isokinetic knee flexion and extension contractions were performed by healthy subjects at five different velocities and at three contraction levels (100%, 75% and 50% of MVC). Joint angle, angular velocity, joint moment and surface EMG of five knee muscles were recorded. Individual calibration values were calculated according to [C.A.M. Doorenbosch, J. Harlaar, A clinically applicable EMG-force model to quantify active stabilization of the knee after a lesion of the anterior cruciate ligament, Clinical Biomechanics 18 (2003) 142-149] for each contraction level. RESULTS: First, the output of the model, calibrated with the 100% MVC was compared to the actually exerted net knee moment at the dynamometer. Normalized root mean square errors were calculated [A.L. Hof, C.A.N. Pronk, J.A. van Best, Comparison between EMG to force processing and kinetic analysis for the calf muscle moment in walking and stepping, Journal of Biomechanics 20 (1987) 167-187] to compare the estimated moments with the actually exerted moments. Mean RMSD errors ranged from 0.06 to 0.21 for extension and from 0.12 to 0.29 for flexion at the 100% trials. Subsequently, the calibration results of the 50% and 75% MVC calibration procedures were used. A standard signal, representing a random EMG level was used as input in the EMG force model, to compare the three models. Paired samples t-tests between the 100% MVC and the 75% MVC and 50% MVC, respectively, showed no significant differences (p>0.05). CONCLUSION: The application of submaximal contractions of larger than 50% MVC is suitable to calibrate a simple EMG to force model for knee extension and flexion. This means that in clinical practice, the EMG to force model can be applied by patients who cannot exert maximal force.     

An EMG-driven musculoskeletal model to estimate muscle forces and knee joint moments in vivo

This paper examined if an electromyography (EMG) driven musculoskeletal model of the human knee could be used to predict knee moments, calculated using inverse dynamics, across a varied range of dynamic contractile conditions. Muscle-tendon lengths and moment arms of 13 muscles crossing the knee joint were determined from joint kinematics using a three-dimensional anatomical model of the lower limb. Muscle activation was determined using a second-order discrete non-linear model using rectified and low-pass filtered EMG as input. A modified Hill-type muscle model was used to calculate individual muscle forces using activation and muscle tendon lengths as inputs. The model was calibrated to six individuals by altering a set of physiologically based parameters using mathematical optimisation to match the net flexion/extension (FE) muscle moment with those measured by inverse dynamics. The model was calibrated for each subject using 5 different tasks, including passive and active FE in an isokinetic dynamometer, running, and cutting manoeuvres recorded using three-dimensional motion analysis. Once calibrated, the model was used to predict the FE moments, estimated via inverse dynamics, from over 200 isokinetic dynamometer, running and sidestepping tasks. The inverse dynamics joint moments were predicted with an average R(2) of 0.91 and mean residual error of approximately 12 Nm. A re-calibration of only the EMG-to-activation parameters revealed FE moments prediction across weeks of similar accuracy. Changing the muscle model to one that is more physiologically correct produced better predictions. The modelling method presented represents a good way to estimate in vivo muscle forces during movement tasks.     

A regression model predicting isometric shoulder muscle activities from arm postures and shoulder joint moments

Tissue overloading is a major contributor to shoulder musculoskeletal injuries. Previous studies attempted to use regression-based methods to predict muscle activities from shoulder kinematics and shoulder kinetics. While a regression-based method can address co-contraction of the antagonist muscles as opposed to the optimization method, most of these regression models were based on limited shoulder postures. The purpose of this study was to develop a set of regression equations to predict the 10th percentile, the median, and the 90th percentile of normalized electromyography (nEMG) activities from shoulder postures and net shoulder moments. Forty participants generated various 3-D shoulder moments at 96 static postures. The nEMG of 16 shoulder muscles was measured and the 3-D net shoulder moment was calculated using a static biomechanical model. A stepwise regression was used to derive the regression equations. The results indicated the measured range of the 3-D shoulder moment in this study was similar to those observed during work requiring light physical capacity. The r² of all the regression equations ranged between 0.228 and 0.818. For the median of the nEMG, the average r² among all 16 muscles was 0.645, and the five muscles with the greatest r² were the three deltoids, supraspinatus, and infraspinatus. The results can be used by practitioners to estimate the range of the shoulder muscle activities given a specific arm posture and net shoulder moment.     

Effects of an anatomically detailed erector spinae model on L4/L5 disc compression and shear

Biomechanical models utilized for analysis of tasks that load the lumbar spine often predict the resultant moment, disc compression and sometimes shear. Usually the extensor muscular and ligament forces of the lumbar spine are assumed to act 5 cm posterior to a disc centre of rotation. This study has re-examined the generation and pathways of muscular force transmission within the extensor musculature. The effects on L4/L5 disc compression and shear estimates of an anatomically and biomechanically justifiable range of tissue moment arms, lines of force and force generating capacity of muscle, input to a computer model, have been determined. Results indicated that L4/L5 compression estimates could be reduced by up to 35% when the output from a more realistic anatomical model of the erector spinae muscle group was compared with that from the frequently reported and simplified single muscle equivalent with a 5 cm moment arm. The shear force estimates could be altered from more than 500 N (L4 tending to shear anteriorly on L5) to less than 200 N with L4 tending to shear posteriorly on L5. Using the combination of input variables considered by the authors to be most feasible to estimate compression, a single 'equivalent' extensor soft tissue moment arm of 7.5 rather than 5 cm would be needed to equate the compression. This simplification of course, does not accommodate the shear force estimate problem.     

Predictions of knee and ankle moments of force in walking from EMG and kinematic data

A deterministic model was developed and validated to calculate instantaneous ankle and knee moments during walking using processed EMG from representative muscles, instantaneous joint angle as a correlate of muscle length and angular velocity as a correlate of muscle velocity, and having available total instantaneous joint moments for derivation of certain model parameters. A linear regression of the moment on specifically processed EMG, recorded while each subject performed cycled isometric calibration contractions, yielded the constants for a basic moment-EMG relationship. Using the resultant moment for optimization, the predicted moment was proportionally augmented for longer muscle lengths and reduced for shorter lengths. Similarly, the predicted moment was reduced for shortening velocities and increased if the muscle was lengthening. The plots of moments predicted using the full model and those calculated from link segment mechanics followed each other quite closely. The range of root mean square errors were: 3.2-9.5 Nm for the ankle and 4.7-13.0 Nm for the knee.     

Lifting over an obstacle: effects of one-handed lifting and hand support on trunk kinematics and low back loading

Mechanical loading of the low back during lifting is a common cause of low back pain. In this study two-handed lifting is compared to one-handed lifting (with and without supporting the upper body with the free hand) while lifting over an obstacle. A 3-D linked segment model was combined with an EMG-assisted trunk muscle model to quantify kinematics and joint loads at the L5S1 joint. Peak total net moments (i.e., the net moment effect of all muscles and soft tissue spanning the joint) were found to be 10+/-3% lower in unsupported one-handed lifting compared to two-handed lifting, and 30+/-8% lower in supported compared to unsupported one-handed lifting. L5S1 joint forces also showed reductions, but not of the same magnitude (18+/-8% and 15+/-10%, respectively, for compression forces, and 15+/-17% and 11+/-14% respectively, for shear forces). Those reductions of low back load were mainly caused by a reduction of trunk and load moment arms relative to the L5S1 joint during peak loading, and, in the case of hand support, by a support force of about 250 N. Stretching one leg backward did not further reduce low back load estimates. Furthermore, one-handed lifting caused an 6+/-8 degrees increase in lateral flexion, a 9+/-5 degrees increase in twist and a 6+/-6 degrees decrease in flexion. Support with the free hand caused a small further increase in lumbar twisting. It is concluded that one-handed lifting, especially with hand support, reduces L5S1 loading but increases asymmetry in movements and moments about the lumbar spine.     

Effect of changes in orientation and position of external loads on trunk muscle activity and kinematics in upright standing

Forces at different heights and orientations are often carried by hands while performing occupational tasks. Trunk muscle activity and spinal loads are likely dependent on not only moments but also the orientation and height of these forces. Here, we measured trunk kinematics and select superficial muscle activity of 12 asymptomatic subjects while supporting forces in hands in upright standing. Magnitude of forces in 5 orientations (−25°, 0°, 25°, 50° and 90°) and 2 heights (20 cm and 40 cm) were adjusted to generate flexion moments of 15, 30 and 45 N m at the L5-S1 disc centre. External forces were of much greater magnitude when applied at lower elevation or oriented upward at 25°. Spinal kinematics remained nearly unchanged in various tasks.Changes in orientation and elevation of external forces substantially influenced the recorded EMG, despite similar trunk posture and identical moments at the L5-S1. Greater EMG activity was overall recorded under larger forces albeit constant moment. Increases in the external moment at the L5-S1 substantially increased EMG in extensor muscles (p < 0.001) but had little effect on abdominals; e.g., mean longissimus EMG for all orientations increased by 38% and 75% as the moment level altered from 15 N m to 30 N m and to 45 N m while that in the rectus abdominus increased only by 2% and 4%, respectively. Under 45 N m moment and as the load orientation altered from 90° to 50°, 25°, 0° and −25°, mean EMG dropped by 3%, 12%, 12% and 1% in back muscles and by 17%, 17%, 19% and 13% in abdominals, respectively. As the load elevation increased from 20 cm to 40 cm, mean EMG under maximum moment decreased by 21% in back muscles and by 17% in abdominals.Due to the lack of EMG recording of deep lumbar muscles, changes in relative shear/compression components and different net moments at cranial discs despite identical moments at the caudal L5-S1 disc, complementary model studies are essential for a better comprehension of neuromuscular strategies in response to alterations in load height and orientation.     

Effects of EMG processing on biomechanical models of muscle joint systems: sensitivity of trunk muscle moments, spinal forces, and stability

Biomechanical models are in use to estimate parameters such as contact forces and stability at various joints. In one class of these models, surface electromyography (EMG) is used to address the problem of mechanical indeterminacy such that individual muscle activation patterns are accounted for. Unfortunately, because of the stochastical properties of EMG signals, EMG based estimates of muscle force suffer from substantial estimation errors. Recent studies have shown that improvements in muscle force estimation can be achieved through adequate EMG processing, specifically whitening and high-pass (HP) filtering of the signals. The aim of this paper is to determine the effect of such processing on outcomes of a biomechanical model of the lumbosacral joint and surrounding musculature. Goodness of fit of estimated muscle moments to net moments and also estimated joint stability significantly increased with increasing cut-off frequencies in HP filtering, whereas no effect on joint contact forces was found. Whitening resulted in moment estimations comparable to those obtained from optimal HP filtering with cut-off frequencies over 250 Hz. Moreover, compared to HP filtering, whitening led to a further increase in estimated joint-stability. Based on theoretical models and on our experimental results, we hypothesize that the processing leads to an increase in pick-up area. This then would explain the improvements from a better balance between deep and superficial motor unit contributions to the signal.     

Criterion validity and between-day reliability of an inertial-sensor-based trunk postural stability test during unstable sitting

IntroductionAdequate neuromuscular control of the lumbar spine is required to prevent lumbar injuries. A trunk postural stability test has been proposed earlier, using a chair wobbling on a central pivot and four springs with adjustable positions to modulate task difficulty. An inertial sensor is fixed on the chair to measure postural sway. The aim of this study is to assess the criterion validity and between-day reliability of the calibration and testing components.MethodsThirty six subjects (with and without low back pain) followed a calibration procedure, four practice trials and three 60-s trials on 2 days. The criterion validity of the inertial sensor was tested against an optoelectronic system and a force platform. The reliability of 38 body sway measures obtained from the inertial sensor angular measures was estimated.ResultsThe inertial sensor led to valid estimates of postural sway. The reliability of the calibration procedure was moderate. Practically no learning effect was detected except for a few body sway measures in patients with CLBP. Three 60-s trials provided acceptable reliability for approximately half of the body sway measures, although this is more difficult to achieve in patients with CLBP.DiscussionThe use of an easy to use inertial sensor led to valid measures of postural sway. A number of body sway measures were identified as reliable tools for individual follow-ups but inter-subject comparisons were anticipated as more difficult when patients with CLBP are involved.     

Muscular resistance to varus and valgus loads at the elbow.

Although the contributions of passive structures to stability of the elbow have been well documented, the role of active muscular resistance of varus and valgus loads at the elbow remains unclear. We hypothesized that muscles: (1) can produce substantial varus and valgus moments about the elbow, and (2) are activated in response to sustained varus and valgus loading of the elbow. To test the first hypothesis, we developed a detailed musculoskeletal model to estimate the varus and valgus moment-generating capacity of the muscles about the elbow. To test the second hypothesis, we measured EMGs from 11 muscles in four subjects during a series of isometric tasks that included flexion, extension, varus, and valgus moments about the elbow. The EMG recordings were used as inputs to the elbow model to estimate the contributions of individual muscles to flexion-extension and varus-valgus moments. Analysis of the model revealed that nearly all of the muscles that cross the elbow are capable of producing varus or valgus moments; the capacity of the muscles to produce varus moment (34 Nm) and valgus moment (35 Nm) is roughly half of the maximum flexion moment (70 Nm). Analysis of the measured EMGs showed that the anconeus was the most significant contributor to valgus moments and the pronator teres was the largest contributor to varus moments. Although our results show that muscles were activated in response to static varus and valgus loads, their activations were modest and were not sufficient to balance the applied load.     

Muscle contributions to recovery from forward loss of balance by stepping

The purpose of this study was to determine the muscular contributions to the stepping phase of recovery from forward loss of balance in 5 young and 5 older adults that were able to recover balance in a single step, and 5 older adults that required multiple steps. Forward loss of balance was achieved by releasing participants from a static forward lean angle. All participants were instructed to attempt to recover balance by taking a rapid single step. A scalable anatomical model consisting of 36 degrees-of-freedom was used to compute kinematics and joint moments from motion capture and force plate data. Forces for 94 muscle actuators were computed using static optimisation and induced acceleration analysis was used to compute individual muscle contributions to net lumbar spine joint, and stepping side hip joint and knee joint accelerations during recovery. Older adults that required multiple recovery steps used a significantly shorter and faster initial recovery step and adopted significantly more trunk flexion throughout recovery compared to the older single steppers. Older multiple steppers also produced significantly more force in the stance side hamstrings, which resulted in significantly higher hamstring induced flexion accelerations at the lumbar spine and extension accelerations at the hip. However since the net joint lumbar spine and hip accelerations remained similar between older multiple steppers and older single steppers, we suggest that the recovery strategy adopted by older multiple steppers was less efficient as well as less effective than for older single steppers.     

Effects of a passive exoskeleton on the mechanical loading of the low back in static holding tasks

With mechanical loading as the main risk factor for LBP in mind, exoskeletons are designed to reduce the load on the back by taking over a part of the required moment. The present study assessed the effect of a passive exoskeleton on back and abdominal muscle activation, hip and lumbar flexion and on the contribution of both the human and the exoskeleton to the L5/S1 net moment, during static bending at five different hand heights. Two configurations of the exoskeleton (LOW & HIGH) differing in angle-torque characteristics were tested. L5/S1 moments generated by the subjects were significantly reduced (15–20% for the most effective type) at all hand heights. LOW generated 4–11 Nm more support than HIGH at 50%, 25% and 0% upright stance hand height and HIGH generated 4–5 Nm more support than LOW at 100% and 75%. Significant reductions (11–57%) in back muscle activity were found compared to WITHOUT for both exoskeletons for some conditions. However, EMG reductions compared to WITHOUT were highly variable across subjects and not always significant. The device allowed for substantial lumbar bending (up to 70°) so that a number of participants showed the flexion-relaxation phenomenon, which prevented further reduction of back EMG by the device and even an increase from 2% to 6% MVC in abdominal activity at 25% hand height. These results indicate that flexion relaxation and its interindividual variation should be considered in future exoskeleton developments.     

Validation of an EMG-driven, graphically based isometric musculoskeletal model of the cervical spine

EMG-driven musculoskeletal modeling is a method in which loading on the active and passive structures of the cervical spine may be investigated. A model of the cervical spine exists; however, it has yet to be criterion validated. Furthermore, neck muscle morphometry in this model was derived from elderly cadavers, threatening model validity. Therefore, the overall aim of this study was to modify and criterion validate this preexisting graphically based musculoskeletal model of the cervical spine. Five male subjects with no neck pain participated in this study. The study consisted of three parts. First, subject-specific neck muscle morphometry data were derived by using magnetic resonance imaging. Second, EMG drive for the model was generated from both surface (Drive 1: N=5) and surface and deep muscles (Drive 2: N=3). Finally, to criterion validate the modified model, net moments predicted by the model were compared against net moments measured by an isokinetic dynamometer in both maximal and submaximal isometric contractions with the head in the neutral posture, 20 deg of flexion, and 35 deg of extension. Neck muscle physiological cross sectional area values were greater in this study when compared to previously reported data. Predictions of neck torque by the model were better in flexion (18.2% coefficient of variation (CV)) when compared to extension (28.5% CV) and using indwelling EMG did not enhance model predictions. There were, however, large variations in predictions when all the contractions were compared. It is our belief that further work needs to be done to improve the validity of the modified EMG-driven neck model examined in this study. A number of factors could potentially improve the model with the most promising probably being optimizing various modeling parameters by using methods established by previous researchers investigating other joints of the body.     

Estimating the L5S1 flexion/extension moment in symmetrical lifting using a simplified ambulatory measurement system

Mechanical loading of the spine has been shown to be an important risk factor for the development of low-back pain. Inertial motion capture (IMC) systems might allow measuring lumbar moments in realistic working conditions, and thus support evaluation of measures to reduce mechanical loading. As the number of sensors limits applicability, the objective of this study was to investigate the effect of the number of sensors on estimates of L5S1 moments.Hand forces, ground reaction forces (GRF) and full-body kinematics were measured using a gold standard (GS) laboratory setup. In the ambulatory setup, hand forces were estimated based on the force plates measured GRF and body kinematics that were measured using (subsets of) an IMC system. Using top-down inverse dynamics, L5S1 flexion/extension moments were calculated.RMSerrors (Nm) were lowest (16.6) with the full set of 17 sensors and increased to 20.5, 22 and 30.6, for 8, 6 and 4 sensors. Absolute errors in peak moments (Nm) ranged from 17.7 to 16.4, 16.9 and 49.3 Nm, for IMC setup’s with 17, 8, 6 and 4 sensors, respectively. When horizontal GRF were neglected for 6 sensors, RMSerrors and peak moment errors decreased from 22 to 17.3 and from 16.9 to 13 Nm, respectively.In conclusion, while reasonable moment estimates can be obtained with 6 sensors, omitting the forearm sensors led to unacceptable errors. Furthermore, vertical GRF information is sufficient to estimate L5S1 moments in lifting.