AI Noise Reduction for Hearing Loss

Overseen By Clinical Research Study Coordinator

Age: 18+

Sex: Any

Trial Phase: Academic

Sponsor: Mayo Clinic

Disqualifiers: None

No Placebo Group All trial participants will receive the active study treatment (no placebo)

Approved in 4 Jurisdictions This treatment is already approved in other countries

Compensation: Varies

Number of Visits: 1

Duration: 4-6 weeks

500 Participants Needed

What You Need to Know Before You Apply

What is the purpose of this trial?

The purpose of this study is to determine if artificial intelligence- based noise reduction can offer objective improvement in hearing and speech perception for hearing impaired patients.

Do I need to stop my current medications for the trial?

The trial information does not specify whether you need to stop taking your current medications.

How is the DEEPSONIC AI chip treatment different from other hearing loss treatments?

The DEEPSONIC AI chip is unique because it uses artificial intelligence to reduce background noise, helping people with hearing loss understand speech better in noisy environments, unlike traditional hearing aids and cochlear implants that struggle in such situations. ^{1 2 3 4 5}

What data supports the effectiveness of the treatment DEEPSONIC™ Deep Sound Optimized Neural Integrated Chip for hearing loss?

Research shows that using deep learning-based noise reduction, like the DEEPSONIC AI chip, can improve speech understanding in noisy environments for cochlear implant users, which suggests it may help people with hearing loss hear better in similar situations. ^{1 4 5 6 7}

Who Is on the Research Team?

Aniket Saoji, PhD AuD

Principal Investigator

Mayo Clinic

Are You a Good Fit for This Trial?

This trial is for individuals with hearing impairment, whether they are new to using hearing aids or currently use them. It also includes those who have a cochlear implant in one ear and use a hearing aid in the other. People already satisfied with their current hearing aids or unable to follow the study protocol may not qualify.

Inclusion Criteria

Patients with hearing impairment not using hearing aids

Cochlear implant patients with a single-sided implant (using a hearing aid in the opposite ear)

Individuals with hearing impairment currently using hearing aids

Timeline for a Trial Participant

Screening

Participants are screened for eligibility to participate in the trial

2-4 weeks

Baseline Testing

Participants undergo baseline testing including single-word and sentence testing to assess hearing and speech perception

1 week

1 visit (in-person)

Treatment

Participants use the Phonak hearing aid with AI-based noise reduction to evaluate improvement in hearing and speech perception

4-6 weeks

Follow-up

Participants are monitored for safety and effectiveness after treatment

4 weeks

What Are the Treatments Tested in This Trial?

Interventions

• DEEPSONIC™ Deep Sound Optimized Neural Integrated Chip

Trial Overview The trial is testing DEEPSONIC™, an AI-based noise reduction technology integrated into Phonak Hearing Aids. The goal is to see if this advanced feature can improve speech understanding and overall hearing in noisy environments compared to standard models.

How Is the Trial Designed?

Group I: Hearing Aid only Experimental Treatment 1 Intervention

Patients with hearing impairment who do not have a Cochlear implant

Group II: Cochlear Implant with Hearing Aid Experimental Treatment 1 Intervention

Patients with hearing impairment who have a single-sided implant (using a hearing aid on the opposite ear)

DEEPSONIC™ Deep Sound Optimized Neural Integrated Chip is already approved in European Union, United States, Canada, Switzerland for the following indications:

Approved in European Union as DEEPSONIC AI chip for:

• Hearing impairment
• Speech perception improvement

Approved in United States as DEEPSONIC AI chip for:

• Hearing impairment
• Speech perception improvement

Approved in Canada as DEEPSONIC AI chip for:

• Hearing impairment
• Speech perception improvement

Approved in Switzerland as DEEPSONIC AI chip for:

• Hearing impairment
• Speech perception improvement

Who Is Running the Clinical Trial?

Mayo Clinic

Lead Sponsor

Trials

3,427

Recruited

3,221,000+

Published Research Related to This Trial

Osseointegrated auditory implants like Baha(®) effectively deliver sound and amplify hearing for patients with conductive or mixed hearing loss and single-sided sensorineural deafness, addressing key challenges in bone conduction hearing.

Recent advancements in technology, such as computer-based fitting software and automatic directional microphones, enhance the hearing performance of Baha users, particularly in noisy environments.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Challenges and recent developments in sound processing for Baha®. Flynn, MC.[2019]

A new deep neural network (DNN) called 'Deep ACE' has been developed to enhance cochlear implant (CI) performance by effectively removing background noise while maintaining the quality of sound coding, which is crucial for users with profound hearing loss.

In tests with eight CI users, Deep ACE significantly improved speech intelligibility scores compared to traditional sound coding strategies, indicating its potential to enhance hearing in noisy environments.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

A Deep Denoising Sound Coding Strategy for Cochlear Implants. Gajecki, T., Zhang, Y., Nogueira, W.[2023]

Researchers developed advanced 'smart' hearing aids and cochlear implants using artificial intelligence to effectively filter out background noise, improving speech understanding in noisy environments.

This technology allows individuals with hearing loss to better engage in conversations in challenging settings, such as busy classrooms or restaurants, enhancing their social interactions and quality of life.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Helping People Hear Better with "Smart" Hearing Devices. Goehring, T., Monaghan, J.[2022]

A direct acoustic cochlear implant successfully improved hearing in a 47-year-old male with severe mixed hearing loss due to otosclerosis, raising his speech recognition score from 10% to 100%.

The restoration of hearing not only enhanced the patient's auditory capabilities but also led to a significant reduction in auditory hallucinations and allowed for a decrease in psychiatric medication, highlighting the link between hearing loss and psychosis.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Successful hearing improvement with direct acoustic stimulation in a patient with schizophrenia. Rebol, J., Kravos, M., Ložar, M.[2018]

The study introduced a new noise reduction system for cochlear implants, called NC+DDAE_T, which uses knowledge transfer technology to reduce the number of parameters needed while maintaining similar performance to the previous NC+DDAE model.

Testing with 10 cochlear implant users showed that the NC+DDAE_T system performed well in noisy conditions, potentially lowering implementation costs and improving the listening experience for users.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Improved Environment-Aware-Based Noise Reduction System for Cochlear Implant Users Based on a Knowledge Transfer Approach: Development and Usability Study. Li, LP., Han, JY., Zheng, WZ., et al.[2021]

A new ear modeling method using acoustic filter theory was developed for a digital hearing aid chip, allowing for real-time feedback and improved programmability.

The fabricated digital hearing aid chip, built using 0.18 microm CMOS technology, is compact (3.1 mm x 1.2 mm) and energy-efficient, consuming less than 120 microamperes, making it suitable for practical use.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

The reference ear modeling method for internally feedback controlled digital hearing aid chip. Kim, S., Lee, SJ., Cho, N., et al.[2020]

A new deep learning algorithm has been developed that effectively suppresses background noise while preserving speech signals, significantly improving speech intelligibility for hearing aid users to levels comparable to those with normal hearing.

This algorithm, trained on a large database of noisy speech, operates in real-time on a single microphone and shows promise for future integration into hearing aids, potentially enhancing the quality of life for millions of people with hearing loss.

NCBI (Pubmed)

pubmed.ncbi.nlm.nih.gov

Restoring speech intelligibility for hearing aid users with deep learning. Diehl, PU., Singer, Y., Zilly, H., et al.[2023]

Citations

1. Switzerland pubmed.ncbi.nlm.nih.gov

Challenges and recent developments in sound processing for Baha®. [2019]

2. United States pubmed.ncbi.nlm.nih.gov

A Deep Denoising Sound Coding Strategy for Cochlear Implants. [2023]

3. Switzerland pubmed.ncbi.nlm.nih.gov

Helping People Hear Better with "Smart" Hearing Devices. [2022]

4. England pubmed.ncbi.nlm.nih.gov

Successful hearing improvement with direct acoustic stimulation in a patient with schizophrenia. [2018]

5. Canada pubmed.ncbi.nlm.nih.gov

Improved Environment-Aware-Based Noise Reduction System for Cochlear Implant Users Based on a Knowledge Transfer Approach: Development and Usability Study. [2021]

6. United States pubmed.ncbi.nlm.nih.gov

The reference ear modeling method for internally feedback controlled digital hearing aid chip. [2020]

7. England pubmed.ncbi.nlm.nih.gov

Restoring speech intelligibility for hearing aid users with deep learning. [2023]