Cross-Species Word Recognition Using SVM-Based Cochlear Implant Coding Strategies

Session Number

Project ID: CMPS 26

Advisor(s)

Dr. Claus-Peter Richter, Northwestern University, Feinberg School of Medicine

Discipline

Computer Science

Start Date

17-4-2024 8:55 AM

End Date

17-4-2024 9:10 AM

Abstract

This study investigates the feasibility of using support vector machine (SVM) algorithms and other advanced classification techniques to decode neural responses from the guinea pig auditory system (inferior colliculus) while aiming to discern a spoken word processed by the animal. Utilizing multi-channel electrodes to record neural response patterns from the inferior colliculus of four guinea pigs, we develop SVM models to infer the uttered words. While our results are preliminary auditory signals. Integrating insights from this approach could inform the optimization of cochlear implant coding strategies for human recipients with hearing disorders. Future research endeavors will further refine and validate these SVM models and collect more neural response data from the guinea pigs, which will, indicating ongoing experimentation, initial findings suggest potential for SVM-based decoding of guinea pig ultimately contribute to advancements in cross-species neural communication and auditory prosthetic technologies.

Share

COinS
 
Apr 17th, 8:55 AM Apr 17th, 9:10 AM

Cross-Species Word Recognition Using SVM-Based Cochlear Implant Coding Strategies

This study investigates the feasibility of using support vector machine (SVM) algorithms and other advanced classification techniques to decode neural responses from the guinea pig auditory system (inferior colliculus) while aiming to discern a spoken word processed by the animal. Utilizing multi-channel electrodes to record neural response patterns from the inferior colliculus of four guinea pigs, we develop SVM models to infer the uttered words. While our results are preliminary auditory signals. Integrating insights from this approach could inform the optimization of cochlear implant coding strategies for human recipients with hearing disorders. Future research endeavors will further refine and validate these SVM models and collect more neural response data from the guinea pigs, which will, indicating ongoing experimentation, initial findings suggest potential for SVM-based decoding of guinea pig ultimately contribute to advancements in cross-species neural communication and auditory prosthetic technologies.