Abstract— Muscular weakness is one of the major impairments limiting motor function following a hemispheric stroke. The objective of this preliminary study was to examine possible motor unit (MU) structural changes in paretic muscle post-stroke as a measure by which to assess neural and/or biomechanical mechanisms of paresis. A surface electromyogram (sEMG) recording and decomposition system was used to record sEMG signals and extract single MU activities from the first dorsal interosseous muscle (FDI) of three hemiparetic stroke survivors. To characterize potential MU structural changes, an estimate of the motor unit action potential (MUAP) amplitude and duration was derived using the spike triggered averaging of the sEMG signal. Our preliminary results reveal MUAPs with systematically smaller amplitude and longer duration in the paretic muscle compared with the contralateral muscle of three tested stroke subjects with varying degrees of motor impairment. The changes in MU properties such as reduced MU size and a reduction in the muscle fiber conduction velocity could contribute at least in part, to muscle weakness post-stroke. The sEMG recording and decomposition system combined with our spike triggered averaging technique has the potential to provide an assessment tool for muscular weakness post-stroke. I. INTRODUCTION Cerebral stroke is a leading cause of disability in the United States [1]. After a stroke injury, muscular weakness is one of the major impairments limiting motor function in stroke survivors [2-4]. Possible mechanisms of weakness include reduced excitatory descending drive, muscle atrophy, and disturbance in the control of the MU pool [5-9]. However, the choice as to which mechanisms should be targeted during therapeutic intervention is still unclear, partly due to the insufficient understanding of relative impact of these particular mechanisms [10-12]. Only a few studies have investigated disturbances of MU structural properties, using parameters such as MU size and muscle fiber conduction velocity in paretic muscles of stroke survivors [5, 13, 14] as a means by which to understand the underlying mechanism of weakness. A reduction in MU size *Research supported by the National Institutes of Health of the USA (Grant #: R24 HD50821-07). Brian Jeon is with the Sensory Motor Performance Program of the Rehabilitation Institute of Chicago (email: bjeon6@johnshopkins.edu) Nina L. Suresh is with the Sensory Motor Performance Program of the Rehabilitation Institute of Chicago (email: n-suresh@northwestern.edu) Aneesha K. Suresh is with the Sensory Motor Performance Program of the Rehabilitation Institute of Chicago (email: aneesha@uchicago.edu) William Z. Rymer is with the Rehabilitation Institute of Chicago and the Department of Physical Medicine and Rehabilitation of Northwestern University (email: w-rymer@northwestern.edu) Xiaogang Hu is with the Sensory Motor Performance Program of the Rehabilitation Institute of Chicago (email: xiaogang.hu@northwestern.edu) could lead to a reduced MU twitch force. Therefore, to achieve any desired overall muscle force, more MUs, possibly more fatigable, need to be activated, which ultimately can lead to muscle weakness. A reduction in fiber conduction velocity could indicate changes in overall distribution of motor unit type as well as changes in the recruitment pattern of motor units. Indeed, earlier studies have shown that the large and fast MUs are more selectively affected compared with small and slow MUs post-stroke [15, 16]. Typically, MU structural changes are assessed using invasive intramuscular EMG recordings or biopsy samplings [13, 16] and the data are collected piecemeal. Thus, a non-invasive and systematic examination of MU structural change post-stroke is necessary to extend our understanding of the neural mechanisms of muscle weakness. A novel sEMG electrode array recording and decomposition method has recently been developed by De Luca and colleagues [17, 18] and it yields a large number of MUs simultaneously over a relatively large force range. Using the results from this sEMG decomposition technique, we have developed an analytical method to examine MUAP properties [19, 20]. Using these novel techniques, the objective of this preliminary study was to examine the change in MU structure to understand the contributions to muscle weakness during voluntary muscle activation post-stroke. II. METHODS A. Subjects Three chronic hemiparetic stroke subjects with varying degrees of weakness (see Table 1) of the extremities contralateral to the cerebral lesion were tested. All participants gave informed consent via protocols approved by the Institutional Review Board at Northwestern University. TABLE I. DEMOGRAPHIC INFORMATION OF STROKE SUBJECTS ID Sex Age Side Time Chedoke FM MVC 1 F 58 R 5 4 38 6.8/16.5 2 F 59 R 23 2 22 8.3/37 3 M 61 R 4 6 63 30/39 Note: Side: paretic side. Time: #years post-stroke. FM: Arm Motor Fugl-Meyer Motor assessment. MVC: MVC ratio: the paretic / contralateral sides B. Experimental Setup The subjects were seated in a Biodex chair with their forearm fixed with a brace and a ring mount to restrain unnecessary movement and minimize the effect of unrecorded muscle activity (Fig. 1A). The elbow of the subject was also comfortably resting on a support during the experiment. The proximal phalanx of the index finger was fixed to a six degrees-of-freedom load cell (ATI, Inc.) while keeping the Motor Unit Structural Change Post Stroke Examined via Surface Electromyography: A Preliminary Report* Brian Jeon, Nina L. Suresh, Aneesha K. Suresh, William Z. Rymer, and Xiaogang Hu 6th Annual International IEEE EMBS Conference on Neural Engineering San Diego, California, 6 - 8 November, 2013 1234 U.S. Government work not protected by U.S. copyright