Search Results

The BHA100 Series Braun® Clear™ Hearing Aid is a self-fitting, air conduction hearing aid, intended to amplify sound for individuals 18 years of age or older with perceived mild to moderate hearing impairment. It is adjusted by the user to meet the user's hearing needs. The device is intended for direct-to-consumer sale and use without the assistance of a hearing health care professional.

The BHA100 Series Braun® Clear™ Hearing Aid is a self-fitting, air conduction hearing aid. It features digital signal processing (16 channel-wide dynamic input compression, 3 channel fast-acting output compression, 16 channel noise reduction, feedback cancellation,) bi-directional microphone with windscreen, volume and program control (environment selection - Quiet, Noisy, Concert, TV), 3 channel equalizer, self-adjustable wire and ear tips, and customization through the Braun® Clear™ Mobile Application. The Braun® ClearCheck™ Hearing Test aims to detect accurate auditory thresholds, partnering with a smart phone application to deliver user-customized sound through the Braun® Clear™ Hearing Aid. These thresholds are used to program the device using a proprietary fitting algorithm.

This document describes a medical device, the BHA100 Series Braun® Clear™ Hearing Aid, and its supporting clinical study. The device is a self-fitting, air conduction hearing aid intended for individuals 18 years or older with perceived mild to moderate hearing impairment, for direct-to-consumer use without professional assistance.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. A table of acceptance criteria and the reported device performance

The document doesn't explicitly present a single table labeled "acceptance criteria" with a clear pass/fail status for each, but the study results are presented against implicit or stated criteria. I've synthesized these from the descriptions of the primary and secondary endpoints.

2. Sample size used for the test set and the data provenance

• Test Set Sample Size:
  • Phase 1 (Quantitative Measures - Audiometric Thresholds & REM): All 80 subjects participated in both self-fitting and audiologist-determined threshold measurements and Real Ear Measurements (REM). This implies 160 ears were tested for the PTA MAD (Figure 2a) and 159 ears for the REM MAD (Figure 3a).
  • Phase 2 (Qualitative Measures - COSI): 80 subjects were randomized (1:1) into two groups: "Audiologist-Fit" (N=40) and "Self-Fit" (N=40). For the COSI analysis, 39-40 subjects per cohort were used (Figure 4a, Figure 4b).
  • Usability Testing: 36 untrained participants.
• Data Provenance: The study was a "prospective, randomized, two-arm, multicenter (4 clinical sites) clinical validation study." The document does not specify the country of origin, but given the FDA submission, it's typically either conducted in the U.S. or under strict international standards acknowledged by the FDA. The nature of the study (prospective) means data was collected specifically for this study under predefined protocols.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

• Phase 1 Ground Truth: The ground truth for audiometric thresholds and Real Ear Measurements (REM) was established by "a licensed audiologist using a calibrated, clinical audiometer." The study was conducted at "4 clinical sites," implying multiple audiologists, though the exact number isn't specified. The qualification is "licensed audiologist."
• Phase 2 Ground Truth: The "Audiologist-Fit" group in Phase 2 served as a comparative arm, where the device was fit by "a licensed audiologist." The ground truth for user satisfaction (COSI) was based on "participant-reported, adjective-scale" responses, comparing the outcomes of self-fitting with audiologist-fitting.
• Usability Testing Ground Truth: The assessment of critical tasks was moderated, but the "ground truth" for usability seems to be based on participants' successful completion of tasks and their subjective assessments, rather than expert judgment of performance against a predefined "correct" interaction.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

The document does not describe an explicit adjudication method (like 2+1 or 3+1 with multiple readers/experts) for establishing ground truth in the clinical performance sections. For audiometric and REM measurements, the comparison was directly between the self-fitting method and a single audiologist's measurement. For the COSI scores, it was a comparison of subject-reported outcomes between two fitting strategies.

5. 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

The study was a "two-arm" clinical validation study comparing self-fitting with audiologist-fitting. It was not an MRMC study in the typical sense of evaluating human reader performance with or without AI assistance. Instead, it directly compared a device's self-fitting capability to a professional fitting, both of which result in a device configuration. The "AI" aspect here is the "proprietary fitting algorithm" in the "Braun® ClearCheck™ Hearing Test" within the mobile application that performs the self-fitting. The study does not quantify how much human audiologists improve with AI assistance, but rather shows that the device's self-fitting capability is non-inferior to traditional audiologist fitting.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, in a sense. Phase 1 of the study specifically evaluated the "self-fitting hearing test of the BHA100 Series Braun® Clear™ Hearing Aid." This "self-fitting" process is driven by the device's algorithm, with the user interacting with the mobile application rather than a human professional for the fitting process. The comparison was then made against audiologist-determined thresholds, effectively evaluating the algorithm's performance in generating fitting parameters.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

• Phase 1 Ground Truth:
  • For audiometric thresholds (PTA): "audiologist executed hearing test [...] using a calibrated, clinical audiometer." This is a form of expert measurement/diagnostics, considered the clinical standard.
  • For Real Ear Measurements (REM): Gold standard clinical measurement (NAL-NL2 targets) performed by a licensed audiologist using calibrated equipment.
• Phase 2 Ground Truth:
  • For COSI scores (user satisfaction): "participant-reported, adjective-scale" responses. This is patient-reported outcomes data.
• Safety: Adverse event reporting and adjudication (clinical site PI determined SAE not device-related).
• Non-clinical: Compliance with established international and national standards (e.g., IEC, ISO, ANSI).

8. The sample size for the training set

The document does not specify a separate "training set" sample size for the device's proprietary fitting algorithm. The information provided heavily focuses on the clinical validation study (test set) for the premarket notification. Device algorithms are typically "trained" on large, diverse datasets before clinical validation. This document only describes the performance evaluation of the already trained and implemented algorithm.

9. How the ground truth for the training set was established

Since the document does not discuss the training phase of the algorithm, it does not specify how the ground truth for any potential training set was established. However, for a self-fitting hearing aid using a proprietary fitting algorithm (Braun® ClearCheck™ Hearing Test), it's highly probable that the algorithm's development and training (if applicable) would have involved extensive audiometric data established by audiologists, similar to the "ground truth" established by the audiologist-executed tests in Phase 1 of this validation study.

Ask a specific question about this device