Evaluating a Noise Reduction Algorithm With Cochlear Implant Users

Not Applicable

Recruiting

• Conditions: Hearing Impairment
• Interventions: Device: SEDA

• Registration Number: NCT05805384
• Lead Sponsor: NYU Langone Health

Brief Summary

In this study, a noise reduction algorithm will be implemented in various listening situations to evaluate its effectiveness in improving speech understanding for cochlear implant users ages 12 and older.

Detailed Description

Not available

Study Timeline

• Study Start: 2023-03-01
• Study First Submitted: 2023-03-23
• Study First Submitted QC: 2023-03-23
• Study First Posted: 2023-04-10
• Status Verified: 2024-12
• Last Update Submitted: 2024-12-30
• Last Update Posted: 2024-12-31
• Primary Completion: 2027-03-01
• Study Completion: 2027-03-01

Recruitment & Eligibility

• Status: RECRUITING
• Sex: All
• Target Recruitment: 60

Inclusion Criteria

• At least 12 years of age.
• Be post-lingually deafened cochlear implant recipients with at least 6 months experience regardless of manufacturer.
• The PI must also deem the child to be developmentally able to complete the experiments.

Exclusion Criteria

• Anyone who is not a cochlear implant user or is aged under 12 years.

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: SINGLE_GROUP

Arm && Interventions

Group	Intervention	Description
Cochlear Implant Users	SEDA	Participants will complete up to 4 experiments evaluating the use of the Speech Enhancement using Dynamic thresholding Approach (SEDA) algorithm. Each testing session will take between 4-6 hours to complete. Testing sessions will continue until the completion of the experiments.

Primary Outcome Measures

Name	Time	Method
Percent of Words Correctly Recognized in Noisy Conditions with Clinical NR	Up to Year 2 (Completion of Experiment 2)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in noisy listening conditions while using a clinical NR algorithm.
Percent of Words Correctly Recognized in Quiet Conditions with Clinical NR	Up to Year 2 (Completion of Experiment 2)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in quite listening conditions while using a clinical NR algorithm.
Percent of Words Correctly Recognized with Very Low-Computation SEDA NR	Up to Year 2 (Completion of Experiment 4)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized with the Very Low-Computation SEDA NR, which is designed to have a reasonable ESTOI value with minimal computational requirements.
Visual Analogue Scale (VAS) Rating of SEDA NR	Up to Year 2 (Completion of Experiment 1.2)	Participants will be presented with two buttons: one button will play a sample with SEDA NR and the other will play a sample without SEDA NR. Listeners will report which sample was preferable by clicking on a horizontal visual analog scale (VAS). VAS ratings will be scored on a scale of -10 to 10 where positive numbers indicate a preference for SEDA NR and negative numbers indicate a preference for not using SEDA NR. The magnitude of the absolute value of the score indicates the strength of the preference.
Percent of Words Correctly Recognized with no SEDA NR	Up to Year 2 (Completion of Experiment 4)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized without SEDA NR.
Percent of Words Correctly Recognized in Noisy Conditions with SEDA NR	Up to Year 2 (Completion of Experiment 1.1)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in noisy listening conditions while using the SEDA Noise Reduction (NR) algorithm.
Percent of Words Correctly Recognized in Quiet Conditions with SEDA NR	Up to Year 2 (Completion of Experiment 1.1)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in quiet listening conditions while using the SEDA NR.
Percent of Words Correctly Recognized in Quiet Conditions without NR	Up to Year 2 (Completion of Experiment 1.1)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in quiet listening conditions with SEDA NR.
Percent of Words Correctly Recognized in Noisy Conditions without NR	Up to Year 2 (Completion of Experiment 1.1)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized in noisy listening conditions without SEDA NR.
Percent of Words Correctly Recognized with Standard SEDA NR	Up to Year 2 (Completion of Experiment 4)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized with Standard SEDA NR.
Percent of Words Correctly Recognized with Low-Computation SEDA NR	Up to Year 2 (Completion of Experiment 4)	Participants will listen to sentences and repeat all the words they heard. This outcome measures the percent of words correctly recognized with the Low-Computation SEDA NR, which is designed to have minimal computation while maintaining an Extended Short-Time Objective Intelligibility Predictor (ESTOI) value similar to the "standard" SEDA implementation

Trial Locations

Locations (1)

Tisch Hospital; 🇺🇸 New York, New York, United States