EMG and ECoG Interfaces for Speech and Spoken Communication - Rehabilitation Institute of Chicago

Skip to Content

Fri, Aug 12

Speaker: Aaron Young

Title: Improving Robustness of Myoelectric Pattern Recognition using Electrode Configuration

Abstract: Myoelectric pattern recognition control can potentially provide upper limb amputees with intuitive control of multiple prosthetic functions. However, the lack of robustness of myoelectric pattern recognition algorithms is a barrier for clinical implementation. One issue that can contribute to poor system performance is electrode shift, which is a change in the location of the electrodes with respect to the underlying muscles that occurs during donning and doffing and daily use. We investigated the effects of interelectrode distance and feature choice on system performance in the presence of electrode shift. Increasing the interelectrode distance from 2 cm to 4 cm significantly (p<0.01) improved classification accuracy in the presence of electrode shifts of up to 2 cm. In a controllability test, increasing the interelectrode distance from 2 cm to 4 cm improved the user’s ability to control a virtual prosthesis in the presence of electrode shift. Use of an autoregressive feature set significantly (p<0.01) reduced sensitivity to electrode shift when compared to use of a traditional time-domain feature set.

Advisor: Dr. Kuiken


Speaker: Emily Mugler

Title: EMG and ECoG Interfaces for Speech and Spoken Communication

Abstract:  The possibility of classifying someone's speech using surface electromyographic (sEMG) technology has great potential to enhance communication in several areas, including high-noise environments such as in airports and highway construction, in communications that require privacy, and in assisting some speech deficits. Several recent research groups have have had limited success using using various electrode quantities, placement locations, classification algorithms, sets of stimuli, and target applications. Our interests are in quantifying the factors that contribute to success and failure, particularly reason some methods do not work. By evaluating such limitations we can identify obstacles to further development. We present experiments that evaluate classification success due to phonemic content, speaking rate, electrode quantity and placement. Finally, we also seek to determine how adding electrocorticographic (ECoG) signals from primary motor cortex can enhance classification success, and present preliminary results.     

Advisor: Dr. Patton