Neuromuscular Control

  •  Brain Control of Hand

    Our long-term goal is to to understand how the brain controls manipulation. To do this it is very useful to record simultaneously from brain, muscles and hand. While this is not possible to do under most circumstances, the ECoG recordings in patients preparing for surgery for epilepsy provides an opportunity to use the ECoG recordings that will be collected already in the course of treatment.Thus, we propose to use the wait time that patients have during the ECoG recordings to record, non-invasively, muscle activity with surface electrodes on forearm and hand muscles, and finger motions and forces with cameras and force sensors. These procedures are all done in addition to the regular treatment, and are not part of it. They simply consist of behavioral tasks that we will ask patients to do in their spare time at the intensive care unit.Therefore, the relationship among ECoG recordings, muscle activity and finger motions and forces is simply one of statistical analysis of their temporal and magnitude correlations. This analysis of correlations will be done off-line. All we will do with the subjects is record these signals simultaneously.Our use of ECoG recordings in patients preparing for surgery for epilepsy is simply a matter of opportunity. We do not know of any other way in which this rich data set can be collected in humans. The fact that these recordings are being collected already in the patients' spare time at the intensive care unit provides a unique and powerful opportunity; it is not a part of their treatment and it does not offer additional risk or discomfort to the patient.

    The ability to conduct this manipulation study on patients already undergoing ECoG recordings allows us for the first time to record activity from the brain, the musculature of the hand/forearm, and the forces exerted by the fingertips, simultaneously. Objectives or Purpose:Investigation study gathering electrographical data (ECoG - recorded as part of patient treatment, EMG, and fingertip forces from force sensors) as patients using their fingers to handle objects. This study is not intended for diagnosis or treatment. The information obtained from this study will measure how well the brain coordinates the action of multiple fingers.

    Neurology patients 18 years of age or older undergoing Video/ECoG monitoring at Keck Hospital of USC, without hand injury or disease, will be asked to enroll. Ten (10) patients will be tested. This study is being conducted on these persons because they are undergoing ECoG recordings as part of their treatment. This provides a unique opportunity to record these data while the participant completes simple manipulation tasks during which EMG from the hand/forearm and fingertip forces are recorded. Study Methodology:The study will utilize a variety of tasks requiring dexterous manipulation of objects and free hand movements. Subjects will perform four tasks for this study: spring compression between index finger and thumb, static grasp, pick up an object and place it at another location, and unconstrained free finger movements. During the tasks, we will measure joint angles of the index finger and thumb and motion of the hand in the workspace. This will be accomplished using motion capture markers and cameras. Up to 20 markers will be attached to the subject using double sided tape while motion data is acquired by four overhead infrared cameras. Surface EMG electrodes will be attached at the proximal part of the forearm and on the hand to capture muscle activity during the tasks.

    Study Endpoints or Outcomes:
    1. Development of cortical map that relates specific areas of cortical activity with endpoint force production.
    2. Brain activity correlation with surface EMG activity and finger forces/motions.
    3. Delays in neural pathways from cortex to endpoint movement/force.

    Index finger and thumb joint angles will be extracted from marker location data captured during the tasks. ECoG data will be used to determine the association of cortical activity with muscle function. ECoG data will be analyzed in the following manner:
    1. Bandpass filtered from about 0.1 Hz to 500 Hz.
    2. Notch filtered at 60 Hz.
    3. Because both ECoG and EMG data consist of numerous channels (each containing anywhere from 16 to 128), the dimension of the data be need to be reduced. This will be accomplished using principle component analysis (PCA) which has the effect of transforming the data set to a lower dimension while maintaining much of the observed variance.
    4. Temporal dependence between ECoG and surface EMG data sets will be determined through correlation.
    5. Processing of the data afterwards is yet undetermined.

    Currently we are considering using statistical methods such as independent component analysis (ICA) to separate sources found within ECoG and EMG data. Furthermore, we may employ Wiener filters to relate surface EMG data to endpoint force magnitude and direction. Additional analysis methods will include,but are not limited to, wavelet analysis, Kalman filtering, least-mean-square adaptive filtering, blind source separation, and blind deconvolution. These techniques establish strength of association and delays of brain activity with hand activity.

    For more information, contact:
    Alexander Reyes
    Study Contact Person
    reyesale@usc.edu

    Dynamic Control of Muscles

    This project is aimed at understanding how the nervous system controls the muscles contributing to the forces generated in isometric knee and elbow extension. Since several muscles, when activated, generate very similar forces, there exist multiple solutions in terms of muscle activation patterns, which all give rise to the same force measured using a force sensor at the shin surface or the hand pressing against a static resistance. Our modeling predicts that the optimal solution to the problem of selecting the appropriate muscle activation pattern for these tasks involves the dynamic switching between the contributing extensor muscles at low MVC percentages, such that fatigue in the muscles is mitigated, since deactivated muscles can recover. In turn, the task performance duration is prolonged as compared to a static activation pattern. Here, in a population of healthy adults, we would like to (i) confirm the presence of switching and (ii) confirm experimentally the MVC percentage level at which the switching is abolished due to a lack of available solution patterns. Finally, we would like to correlated the frequency of activation pattern changes to the reduction in MVC force under fatigue: higher switching frequencies should lead to a decreased rate in MVC force reduction.

    For more information, contact:
    Emily Lawrence
    Study Coordinator
    ellawren@usc.edu

    Control of Hand Muscles

    In this study we are investigating how the nervous system deals with muscular redundancy in the human hand. Muscular redundancy refers to the phenomenon that the hand has more muscles than directions in which the parts of the hands can be moved. Because of redundancy, an infinite number of different muscle activation patterns generate the same forces applied by the fingertips to manipulated objects. We would also like to determine the how these changes in neural strategies occur in the brain through electroencephalography.

    We hypothesize that the nervous system can take advantage of this redundancy to, for example, avoid fatigue in muscles or accommodate task changes, by switching between activation patterns that all generate the same forces. We plan to conduct this study with healthy adults, whom we ask to perform multifinger manipulation tasks with the thumb, index, and/or middle finger (e.g. pick up, hold, rotate and transport different objects, tap and hold, etc.). At the same time, we record the activation patterns of a relevant subset of the 21 muscles that actuate these fingers, using, whenever appropriate, surface EMG or fine-wire EMG, which is a means to measure the level of muscle activation. In addition, we will record from non-invasive EEG scalp electrodes to determine the pattern of cortical activations during static and dynamic manipulation. During the tasks, we will measure joint angles of the index finger and thumb and motion of the hand in the workspace. This will be accomplished using motion capture markers and cameras. Up to 20 markers will be attached to the subject using double sided tape while motion data is acquired by overhead infrared cameras.

    For our analysis, we will use processing techniques such as computing the relative contribution of each muscle at any given time, corticomuscular coherence, temporal correlations and frequency analysis. This information will allow us to estimate the neural commands and strategies for controlling the muscles which actuate the hand.

    For more information, contact:
    Emily Lawrence
    Study Coordinator
    ellawren@usc.edu

    Multi-finger Control Strategies


    This study is about learning how people use their fingers to handle objects.

    You may be eligible if you:

    • Are over the age of 18 
    • Have no history of hand injury or disease
    • Have no history of neurological disorders

    If you agree to participate, the testing will take no more than 45 minutes. We will take measurements of your hands, including finger and pinch strength. We will ask you to reach for, grasp, and manipulate an object with your fingers while we measure the position of your fingers. In some tests we will measure your muscle activity.

    For more information, contact:
    Emily Lawrence
    Study Coordinator
    ellawren@usc.edu

    HS protocol # HS-07-00521
    Grant #: EFRI-COPN 0836042, NIH Grants AR050520 & AR052345
    Sponsor: National Science Foundation, National Institute of Health

    Measuring Natural Movement

     Recreational activity in sports is an essential element of healthy living. However, such activities carry the risk of injury. Clinically, it is difficult to objectively assess lingering movement impairments during natural sports related tasks (i.e. running, full body changes in movement direction or throwing,) in the real world (i.e., outside of the controlled and constrained environment of the clinical tests). This is largely because the clinical gold standard motion analysis tools are not practical in daylight, expensive, and time consuming. As a result, we lack reliable outcome measures for complex sports related tasks as performed in the field or on the court.The purpose of this study is to establish typical coordination patterns between segments in adults during recreational and athletic sports tasks. Recent advances in low-cost wireless technology make it possible to collect acceleration data synchronously from multiple segments during dynamic tasks in the outdoors sports environment. An understanding of how these data relate to typical and impaired movement is needed for the development of objective, low-cost clinical measures of movement during recreational and athletic sports tasks. Otherwise healthy, recreationally active adults between the ages of 18 and 50 will be recruited. Body segment accelerations will be collected using a Bluetooth Data Acquisition system with accelerometer motion sensors during the performance of the target recreational or sports activity. Dynamical systems analysis will be used to characterize the multi-dimensional nature of the movement task.

    For more information, contact:
    Emily Lawrence
    Study Coordinator
    ellawren@usc.edu