RUcore Resource Object
RUcore Resource Object
PDF-1(1.43 MB)
TitleNeural correlates of elbow joint kinematic variability
NameNatarajan, Gautam Siddarth (author), Craelius, William (chair), Shoane, George (internal member), Kim, Nam-Hun (outside member), Rutgers University, Graduate School - New Brunswick,
Degree Date2009-10
Date Created2009
SubjectBiomedical Engineering,
Human mechanics,
Joints--Range of motion,
Elbow--Movements
DescriptionA fundamental tenet of motor control is that point-to-point reaching motions follow an approximately straight line trajectory with a bell-shaped velocity profile. However, these abstractions are not universally observed. Previous work in our lab revealed that most spatiotemporal elbow trajectories do not necessarily conform to a straight-line, which is believed to be ‘natural’ human motion. Instead, spatiotemporal trajectories are best characterized by a small set of simple, analytic functions including both linear and non-linear waveforms. Here, I suggest that the differences observed in elbow kinematics are a direct consequence of varying motor planning, which is represented by the electromyography (EMG).
Fourteen healthy subjects were asked to perform several self-paced, untargeted elbow articulations that maximize smoothness within a comfortable range of motion; EMG of the biceps and kinematic traces were recorded simultaneously. Kinematic traces were modeled by a set of simple, monotonic functions, while EMG traces were reconstructed by parabolic waveforms, via a parameterized curve-fitting method. EMG traces (r-EMG) and their parabolic reconstructions (p-EMG) were used independently to predict adherence to each of 3 kinematic types. It was hypothesized that the p-EMG and r-EMG from the kinematic adherence group (r2 > 0.9979) would exhibit statistical and parametric differences from the departure group (r2 < 0.9951) of the same kinematic type.
Both r-EMG and p-EMG were useful in predicting adherence to global kinematic morphology with high sensitivity and specificity across subjects. Coupling the substantial predictive value and the similar information content (87.80% of the decisions were identical) of both EMG modalities implied that p-EMG can be used as a simple, informative approximant of r-EMG of the biceps during self-paced, untargeted elbow flexions. The features selected for classification were robust across subjects along with the predictive value, suggesting there is degeneracy in the neural command. Degeneracy in the neural command matches the widespread observation of highly stereotyped kinematics. Elbow trajectories most commonly adhered to sigmoid morphology. Future work should develop a comprehensive depiction of the neural control of voluntary movements.
Fourteen healthy subjects were asked to perform several self-paced, untargeted elbow articulations that maximize smoothness within a comfortable range of motion; EMG of the biceps and kinematic traces were recorded simultaneously. Kinematic traces were modeled by a set of simple, monotonic functions, while EMG traces were reconstructed by parabolic waveforms, via a parameterized curve-fitting method. EMG traces (r-EMG) and their parabolic reconstructions (p-EMG) were used independently to predict adherence to each of 3 kinematic types. It was hypothesized that the p-EMG and r-EMG from the kinematic adherence group (r2 > 0.9979) would exhibit statistical and parametric differences from the departure group (r2 < 0.9951) of the same kinematic type.
Both r-EMG and p-EMG were useful in predicting adherence to global kinematic morphology with high sensitivity and specificity across subjects. Coupling the substantial predictive value and the similar information content (87.80% of the decisions were identical) of both EMG modalities implied that p-EMG can be used as a simple, informative approximant of r-EMG of the biceps during self-paced, untargeted elbow flexions. The features selected for classification were robust across subjects along with the predictive value, suggesting there is degeneracy in the neural command. Degeneracy in the neural command matches the widespread observation of highly stereotyped kinematics. Elbow trajectories most commonly adhered to sigmoid morphology. Future work should develop a comprehensive depiction of the neural control of voluntary movements.
NoteM.S.
NoteIncludes bibliographical references (p. 44-45)
Noteby Gautam Siddarth Natarajan
Genretheses
Persistent URLhttp://hdl.rutgers.edu/1782.2/rucore10001600001.ETD.000051879
Languageeng
CollectionGraduate School - New Brunswick Electronic Theses and Dissertations
Organization NameRutgers, The State University of New Jersey
RightsThe author owns the copyright to this work