Measuring and modelling arm dynamics to support studies into reducing tremor in individuals with multiple sclerosis

Abstract

This paper reports on work carried out to date, and future directions for, a project that intends to measure and control tremor in individuals with multiple sclerosis (MS). The project will use computer modelling to simulate the movements involved in upper limb intention tremor, as exhibited by individuals with MS.The models will be incorporated into a final system that will interpret measurements from sets of microelectromechanical systems (MEMS) sensors positioned over the surfaces of the upper limb to provide position, orientation and movement information. These data will be combined with gross muscle activity data, extracted from surface electromyogram (EMG) measurements, to create a fully dynamic model of the arm movements in real time. This model will be used to develop methods of attenuating the tremor, firstly at the elbow joint, without adversely affecting the underlying intended movement, by providing real-time feedback control of appropriate force input devices. The estimates of movement and the forces involved will also provide clinicians with a useful tool in the quantification of upper limb tremor.

Measuring Intention Tremor in Multiple Sclerosis Using Inertial Measurement Unit (Imu) Devices

“…Several commercial systems that can provide 3D orientations in this manner are discussed in Ketteringham, 2010. The sensors used in this work were chosen for their ability to provide 3D orientation measurements in real-time, in an easily applied, unobtrusive system.…”

“…Further analyses can reveal the frequency content of the movements. The joint angles, angular rates and angular accelerations have also been used in inverse dynamics models, created in the Matlab ® SimMechanics™ toolbox, for estimating joint torques during the movements, as described in Ketteringham, 2010 andKetteringham, et al, 2008. The movement measurement system therefore provides a useful tool for measuring tremor levels and movement capabilities in individuals during clinical assessments. It allows quantitative assessment of the efficacy of other treatments in a less subjective manner and in far more detail than is possible with visual assessment alone.…”

“…These joint range of movement, eyesight and muscle strength tests were carried out to ensure that the subjects could comfortably carry out the tasks required during the recorded movements. Further details are given in Ketteringham, 2010. The Fahn test consisted of a tremor assessment by observation while the subject held a pose, to assess postural tremor, and performed a finger-tonose test, to assess kinetic and intention tremor. The subject was asked to hold a pose and perform these actions while a clinician observed the resulting movements.…”

“…Body segment dimensions were measured before attachment of the sensors, as described in Ketteringham, 2010.…”

“…This can give inaccurate results when analysing the resulting rectified data for frequency information (as an example, a rectified and non-rectified sine wave contain different frequency components). A method was created to negate portions of the resulting data, in order to return it to being a waveform that represented the tremor movements alternating to "either side" of the lower frequency data, as discussed in Ketteringham, 2010. Windowed power spectra were calculated at regular intervals through the resulting tremor magnitude data. The power spectra were calculated at time increments of 0.3 s, with a time window size of 1.3 s. This window size was found to be optimal in terms of producing good frequency and time resolution; a shorter window was more responsive to rapid changes in movement while a longer time window gave better differentiation of the frequencies contained in the data.…”