Official websites use .gov

A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS

A lock ( 🔒 ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

Need Help

The Development and Testing of Cochlear Implant Strategies Using Machine Learning

Award Number: R21DC024042
ORGANIZATION: NATIONAL INSTITUTE ON DEAFNESS AND OTHER COMMUNICATION DISORDERS
OPDIV: NIH
AWARD CLASS: DISCRETIONARY
AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)
PERIOD OF PERFORMANCE START DATE: 08/05/2026
PERIOD OF PERFORMANCE END DATE: 07/31/2028

Group Awards By:

View Award Description

The Development and Testing of Cochlear Implant Strategies Using Machine Learning - PROJECT SUMMARY Millions of people around the world are categorized as deaf, with hearing loss so severe that it cannot be treated with a hearing aid. For such individuals, the treatment of choice is a cochlear implant. Cochlear implants bypass the ear to electrically stimulate the auditory nerve, causing the perception of sound. Cochlear implants are one of the great success stories in biomedical engineering, but nonetheless fail to fully restore normal hearing in deaf individuals. Although there has been substantial progress in enabling speech comprehension, users typically struggle to understand speech in environments with background noise or competing sound sources. These difficulties in hearing often lead to frustration and social isolation. The behavioral performance of cochlear implant users is likely to be limited in part by suboptimalities in the strategy by which sound is translated into electrical stimulation. There is thus a critical need to explore alternative stimulation strategies. However, large-scale exploration is currently limited by two factors. First, behavioral outcomes in individual human users improve slowly over time, and do not reach asymptotic performance until many months of continuous use. Second, behavioral outcomes are highly variable across individuals. As a result, the only way to know whether an alternative strategy would produce better behavioral performance would be to have a large group of human participants live with the strategy for many months, which is impractical for large numbers of alternative strategies. Models that can predict the auditory behavior that is possible given different cochlear implant strategies could play a crucial role in this process by enabling large-scale screening of candidate strategies. The goal of the proposed research is to develop such models using machine learning and to use them to evaluate the best-case scenario outcomes for a set of alternative cochlear implant stimulation strategies. We propose to build artificial neural network models that recognize speech, localize sounds, discriminate pitch, and attend to particular sound sources given simulated auditory nerve responses from cochlear implant stimulation. The models will link electrical stimulation of the auditory nerve with auditory behavior, providing an upper bound on how well tasks can be performed with current and alternative implant strategies. We will then use the models to evaluate recently proposed alternative stimulation strategies and to derive stimulation strategies that are optimized for behavioral performance. The goal is to arrive at a small set of promising alternative strategies that will be strong candidates for clinical trials in humans and that would hopefully enable better prosthetic hearing.


Issue Date FY	Funding FY	Legal Entity Name	Legal Entity Address	Legal Entity City	Legal Entity State	Legal Entity Zip Code	Legal Entity COUNTY	Legal Entity COUNTRY	Assistance Listing	Award Code	Budget Year	Action Date	Action Type	Action Amount

Issue Date FY: 2026 ( Subtotal = $421,839 )
2026	2026	MASSACHUSETTS INSTITUTE OF TECHNOLOGY	77 MASSACHUSETTS AVE	CAMBRIDGE	MA	02139	MIDDLESEX	USA	Research Related to Deafness and Communication Disorders	000	1	8/7/2026	NEW	$421,839
														Subtotal = $421,839

Grand Total All Awards = $421,839

Top

All Categories

About

Search

Reports

Data Submission

Award Information

The Development and Testing of Cochlear Implant Strategies Using Machine Learning

Award Number: R21DC024042

ORGANIZATION: NATIONAL INSTITUTE ON DEAFNESS AND OTHER COMMUNICATION DISORDERS

OPDIV: NIH

AWARD CLASS: DISCRETIONARY

AWARD ACTIVITY TYPE: SCIENTIFIC/HEALTH RESEARCH (INCLUDES SURVEYS)

PERIOD OF PERFORMANCE START DATE: 08/05/2026

PERIOD OF PERFORMANCE END DATE: 07/31/2028

Federal Websites

Department of Health & Human Services

HHS Operating Divisions

HHS Staff Divisions

Download A Document Viewer