(149 days)
The Soundwave Hearing Sontro® Self-Fitting OTC Hearing Aids are intended to amplify sounds for use by individuals 18 years and older with perceived mild to moderate hearing impairment. The device is intended for over-the-counter sale and use without the assistance of a hearing care professional.
Per 21 CFR 874.3325, a self-fitting wireless air-conduction hearing aid is a wearable soundamplifying device intended to compensate for impaired hearing. It incorporates technology, including software, that allows users to program their hearing aids. This technology integrates user input with a self-fitting strategy, enabling users to independently derive and customize their hearing aid fitting and settings. This is an over-the-counter hearing aid.
The Soundwave Hearing, Sontro® OTC (over-the-counter) Hearing Aids are self-fitting wireless air conduction hearing aids consisting of the hardware and software: the device uses the otoTune® app and accessories supplied in the carton. Each hearing aid in the pair functions and interacts with the otoTune® app independently and as a system.
A disposable size 312 zinc-air battery powers the hearing aids. The hearing aids incorporate microphones for audio input, and sound is delivered to the ear via a receiver in the canal that can be coupled with ear domes. The hearing aids are controlled via onboard button controls and wirelessly via the otoTune® app (iOS and Android). The controls allow the user to configure parameters, settings, and listening modes.
Here's a breakdown of the acceptance criteria and the study details for the Sontro® OTC Hearing Aids (AI), based on the provided document:
Acceptance Criteria and Device Performance
| Acceptance Criteria (Endpoint) | Reported Device Performance | Non-inferiority Criterion Met? |
|---|---|---|
| Primary Effectiveness: | ||
| SELF-FIT REAR equivalent to CLINIC-FIT NAL/NL2 prescription REAR targets (in-lab) | Average REAR differences (CLINIC-FIT - SELF-FIT) for both ears: 0.17 dB. 95% CI: (-0.44, 0.78). | Yes (CI within +/- 5 dB) |
| SELF-FIT APHAB global score non-inferior to CLINIC-FIT APHAB global scores (wear-time field trial) | Average APHABglobal differences (CLINIC-FIT - SELF-FIT): 1.93. 95% CI: (-4.35, 8.21). | Yes (CI upper bound -1.5 dB) |
| SELF-FIT QuickSIN equivalent to CLINIC-FIT QuickSIN (wear-time field trial) | Average QuickSIN SNR loss differences (CLINIC-FIT - SELF-FIT): -0.06 dB. 95% CI: (-1.47, 1.36). | Yes (CI lower bound > -1.5 dB) |
| REAR for FINAL-FIT equivalent to REAR for AFTERTRIAL-FIT (wear-time field trial) | Average REAR differences (AFTERTRIAL-FIT - FINAL-FIT) for both ears: -0.22 dB. 95% CI: (-0.87, 0.42). | Yes (CI within +/- 5 dB) |
| Secondary Reliability: | ||
| SELF-FIT QuickSIN (A) equivalent to SELF-FIT QuickSIN (B) (in-lab) | Average QuickSIN SNR loss differences (SELF-FIT A - SELF-FIT B): 0.28 dB. 95% CI: (-0.48, 1.03). | Yes (CI lower bound > -1.5 dB) |
Study Details
-
Sample Size used for the test set and the data provenance:
- Sample Size: 29 participants were enrolled in the study (13 men, 16 women). 17 subjects withdrew due to poor compliance or personal reasons, meaning the final analysis was likely based on 12 participants.
- Data Provenance: The document does not explicitly state the country of origin but implies a single clinical study setting. It was a prospective study involving participants with perceived mild to moderate hearing impairment.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- The "ground truth" for the test set in this context refers to the CLINIC-FIT measurements. These were established by an "audiology-best-practices hearing aid fitting," which implies fitting by a qualified audiologist. The document does not specify the exact number of audiologists or their years of experience but indicates adherence to "audiology-best-practices."
-
Adjudication method for the test set:
- The clinical study compared the device's self-fitting method (SELF-FIT) against a clinician-fitted method (CLINIC-FIT). There isn't an explicit "adjudication method" described in the sense of multiple experts reviewing cases. Instead, the "ground truth" (CLINIC-FIT) was established by standard audiological best practices. The comparison was statistical, assessing non-inferiority between the SELF-FIT and CLINIC-FIT results.
-
If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance:
- No, a traditional MRMC study comparing human readers with and without AI assistance was not explicitly stated or conducted as described. This study focuses on the effectiveness of a self-fitting hearing aid compared to a clinician-fitted hearing aid, not on AI assistance for human "readers" of medical images or data.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Yes, the "SELF-FIT" arm of the study represents the performance of the device's self-fitting algorithm without direct human (audiologist) intervention in the fitting process. The user themselves interacts with the app to perform the fitting. However, it's important to note that the user is "in the loop" for the self-fitting process, but an audiologist is not. So while not a fully "algorithm-only" standalone per se, it is standalone clinician-free fitting.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- The ground truth for comparison ("reference standard") was expert-established audiological measurements and patient-reported outcomes, specifically:
- Real-Ear Aided Response (REAR) based on audiology-best-practices NAL/NL2 prescription, representing an objective acoustic measure.
- Abbreviated Profile of Hearing Aid Benefit (APHAB) global scores, representing subjective patient outcomes.
- QuickSIN test results, representing an objective measure of speech recognition in noise.
- The ground truth for comparison ("reference standard") was expert-established audiological measurements and patient-reported outcomes, specifically:
-
The sample size for the training set:
- The document does not provide information regarding a specific training set size for the self-fitting algorithm. The focus of this submission is the clinical validation of the device itself, which incorporates the self-fitting technology. The development and training data for the underlying algorithm are not detailed here.
-
How the ground truth for the training set was established:
- As there is no information on a specific training set in the provided text, there is also no information on how the ground truth for such a set was established. It can be inferred that the NAL-NL2 fitting algorithm, which the device utilizes, is a widely validated and established prescriptive method in audiology, derived from extensive research and clinical data over time, rather than a specific "training set" for this particular device submission.
§ 874.3325 Self-fitting air-conduction hearing aid.
(a)
Identification. A self-fitting air-conduction hearing aid is a wearable sound amplifying device that is intended to compensate for impaired hearing and incorporates technology, including software, that allows users to program their hearing aids. This technology integrates user input with a self-fitting strategy and enables users to independently derive and customize their hearing aid fitting and settings. A self-fitting air-conduction hearing aid is subject to the requirements in § 800.30 or § 801.422 of this chapter, as applicable.(b)
Classification. Class II (special controls). The special controls for this device are:(1) Clinical data must evaluate the effectiveness of the self-fitting strategy.
(2) Electroacoustic parameters, including maximum output limits, distortion levels, self-generated noise levels, latency, and frequency response, must be specified and tested.
(3) Performance data must demonstrate the electromagnetic compatibility (EMC), electrical safety, and thermal safety of the device.
(4) Software verification, validation, and hazard analysis must be performed.
(5) If the device incorporates wireless technology:
(i) Performance testing must validate safety of exposure to non-ionizing radiation; and
(ii) Performance data must validate wireless technology functions.
(6) Usability testing must demonstrate that users can correctly use the device as intended under anticipated conditions of use.