Short communication Assessment of spatio-temporal gait parameters using inertial measurement units in neurological populations Patrick Esser a, *, Helen Dawes a,b , Johnny Collett a , Max G. Feltham a , Ken Howells a a Movement Science Group, School of Life Sciences, Oxford Brookes University, Oxford, United Kingdom b Dept of Clinical Neurology, University of Oxford, Oxford, United Kingdom 1. Introduction Neurological populations such as Parkinson’s disease (PD), muscular dystrophy (MD), motor neuron disease (MND) and stroke survivors make up about three million individuals in the UK [1]. Maintaining mobility is a key concern for people with these conditions [2]. Objective measurement of mobility and gait can be a critical marker for clinicians to monitor disease progression [3]. Not all clinicians have access to a gait laboratory but inertial measurement units (IMU), can be used during standard clinical tests such as the 10 m or 6 min walking tests [4], to provide information on underlying gait performance which can be used to direct rehabilitation [2]. Centre of mass (CoM) motion during walking can be measured by IMU, which combine gyroscope, accelerometer and magnetometer data to provide accurate vertical acceleration, speed and position measurements in the global frame [5]. This data can then be used to determine accurate temporal gait parameters (i.e. step time), however spatial parameters (i.e. step length) need to be estimated utilizing mathematical equations such as (1) d ¼ g ð2 ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 2lh  h 2 q Þ (1) where (d) is the step length, (h) is the CoM vertical excursion, (l) is the leg length and (g) is the correction factor [6,7]. This model is suitable for routine clinical use, because it requires only one additional measurement (leg length). Previous research [8] has shown that in typically developed adults (TDA) the model underestimates step length by 25%, because the anterior–posterior movement during double stance is not included [6]. Therefore, a correction factor of 1.25 is required to estimate step length with the model in TDA [8]. However, it is unclear if the model accurately estimates the step length in populations where gait is affected by a neurological condition [9]. Considering the utility of measuring gait using IMUs in individuals with neurological conditions, the correction factor should now be explored. Previous investigations have shown that Gait & Posture 34 (2011) 558–560 A R T I C L E I N F O Article history: Received 30 November 2010 Received in revised form 20 June 2011 Accepted 26 June 2011 Keywords: IMU Centre of mass Accelerometry Pendulum model A B S T R A C T Laboratory based gait analysis techniques are expensive, time consuming and require technical expertise. Inertial measurement units can directly measure temporal parameters and in combination with gait models may provide a solution to obtain spatial gait measurements within daily clinical assessments. However it is not known if a model and standard correction factor determined by Zijlstra and Hof [8] to estimate step and stride length parameters in typically developed adults (TDA) can be accurately used in neurologically impaired gaits. This research estimated the stride length over two 10 m walks at self selected walking speed in people with neurological conditions, using a previously established model and correction factor for TDA. The relation of the correction factor to walking speed was explored. We recruited TDA (n = 10) and participants with Parkinson’s disease (PD; n = 24), muscular dystrophy (MD; n = 13), motor neuron disease (MND; n = 7) and stroke survivors (n = 18) for the study who twice walked 10 m at a self-selected pace. Stride length correction factors, for TDA (1.25  0.01), PD (1.25  0.03), and MD (1.21  0.08) (p = 0.833 and p = 0.242) were the same as previously reported in TDA (Zijlstra and Hof [8]). Correction factors for stroke (1.17  0.42) and MND (1.10  0.08) were different (p < 0.01 and p = 0.028 respectively). However there was a high level of variability for correction factors within groups, which did not relate to walking speed. Our findings support that correction factors should be determined for each individual to estimate average step/stride length in patients suffering from a neurological condition. ß 2011 Elsevier B.V. All rights reserved. * Corresponding author at: Movement Science Group, Oxford Brookes University, Gipsy Lane Campus, Headington, Oxford, OX3 7QF, United Kingdom. Tel.: +44 019865 483 272. E-mail address: pesser@brookes.ac.uk (P. Esser). Contents lists available at ScienceDirect Gait & Posture jo u rn al h om ep age: ww w.els evier.c o m/lo c ate/g aitp os t 0966-6362/$ – see front matter ß 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.gaitpost.2011.06.018