K Number
K130790
Device Name
LYRIC2
Manufacturer
Date Cleared
2013-06-27

(97 days)

Product Code
Regulation Number
874.3300
Panel
EN
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Lyric 2 hearing aid is a disposable, extended-wear air conduction hearing aid, designed to be used and worn by hearing-impaired persons. Hearing assistance is achieved through amplification of sound pressure waves, which are transmitted to the external ear canal via air conduction. The hearing aid is placed in the ear canal by an appropriately trained ENT Physician, Audiologist or Hearing Aid Dispenser and can remain in the ear canal for up to 4 months or until the battery is depleted. Upon device removal the hearing aid is discarded.

Device Description

Lyric2 is a disposable, programmable, deep-canal, extended-wear, air conduction hearing device, It uses an analog, digitally programmable WDRC ("Wide Dynamic Range Compression") circuit with very low power consumption. Hearing assistance is achieved through amplification of sound pressure waves, which are transmitted to the external ear canal via air conduction. The amplification characteristics are contained in digitally programmable memory, and adjustment of device parameters is achieved through the proprietary Fitting System and Software.

The Lyric2 consists of:

  • A programmable application specific integrated circuit ("ASIC") for analog signal processing. . An analog microphone receives sound waves that will be amplified and transmitted to the speaker.
  • A custom made built-in battery to power the amplification circuit. .
  • A magnetic sensor to program device parameters and to remotely control user settings (On/Off/Sleep/Volume).
  • Mechanical components to protect the device from cerumen. .
  • A removal loop to allow the user and/or professional to remove the device. .
  • . Seals that allow the device to fit comfortably in the ear canal

Accessories available for Lyric2:

  • User Remote Control
  • . Device sizer to determine best device size
  • Length sizer to determine insertion depth .
  • . An insertion tool for insertion and removal of device and device sizers
  • A dedicated software and programming interface to program the device with customer · ● specific settings
AI/ML Overview

The Phonak's Lyric2 is a hearing aid, so the acceptance criteria would revolve around its acoustic performance, safety, and comfort, as well as its functional equivalence to the predicate device.

Here's an analysis of the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria CategorySpecific Criteria (Inferred)Reported Device PerformanceComments
Acoustic PerformanceCompliance with ANSI S3.22:2009 "Specification of Hearing Aid Characteristics""The device met all applicable specifications developed by the company in accordance with design input specifications."This indicates that all acoustic performance metrics (e.g., gain, frequency response, distortion) were likely within predefined limits, potentially established by either the standard itself or specific design input requirements that align with the standard.
Functional EquivalenceComparable performance to the cleared Lyric (predicate device) in various clinical assessments.Confirmed "equivalent performance of the modified Lyric compared to the cleared Lyric." This included assessment for:
  • Appropriate device sizing
  • Clinical assessments of gain
  • Speech understanding
  • Sound quality
  • Length of device wear
  • Patient ratings of comfort
  • Proportion of eligible population meeting sizing parameters
  • Post-removal health of the ear
  • Usability evaluations by patients and audiologists | This is a broad claim of equivalence across multiple functional and clinical aspects. The details of "comparable" would be defined in the study protocol. |
    | Safety | Biocompatibility, electrical safety, and electromagnetic compatibility. | "Biocompatibility testing, electrical safety, and electromagnetic compatibility testing, has confirmed the equivalent performance of the modified Lyric compared to the cleared Lyric." | This indicates the device meets established safety standards and is comparable to the predicate for these aspects. |

2. Sample Size Used for the Test Set and Data Provenance

  • Sample Size: The document does not explicitly state the specific sample size(s) for the clinical testing. It refers to "clinical testing" and "post-market clinical testing" but doesn't quantify the number of participants.
  • Data Provenance: The document does not specify the country of origin. It implicitly describes a prospective study ("Post-market clinical testing also confirmed the comparable performance..."). The phrase "Post-market clinical testing" suggests data collected after the original Lyric (predicate device) was on the market, but this specific study was for the Lyric2.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

  • Number of Experts: Not specified.
  • Qualifications of Experts: Not specified. The evaluations involved "patients and audiologists." While audiologists are qualified professionals, the number and specific experience of those involved in establishing "ground truth" or performing assessments are not detailed.

4. Adjudication Method for the Test Set

  • The document does not describe any adjudication method. Given the nature of the device (hearing aid) and the stated assessments (gain, speech understanding, comfort, etc.), it's more likely that direct measurements and patient/audiologist feedback were collected and potentially aggregated, rather than a consensus-based adjudication of a subjective "ground truth" in the typical sense (e.g., image interpretation).

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

  • No, an MRMC comparative effectiveness study, in the typical sense for diagnostic devices, was likely not done. This type of study (comparing human readers with and without AI assistance on specific cases) is relevant for AI algorithms that aid in interpretation or diagnosis. For a hearing aid, the "effectiveness" is measured by direct device performance and patient outcomes, not by how humans interpret findings with or without the device's "AI."
  • The study did compare the Lyric2 to the predicate Lyric (without AI, as AI in hearing aids was not as prevalent then) across various clinical parameters, but not in an MRMC paradigm comparing human reading performance.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

  • Yes, performance characteristics of the device itself were evaluated in a standalone manner. "The performance characteristics of the Lyric Hearing Aid have been evaluated in accordance with ANSI S3.22:2009, 'Specification of Hearing Aid Characteristics.'" This standard involves direct measurement of the device's acoustic output and characteristics, independent of a human ear. This represents the "algorithm only" or device-only performance.

7. The Type of Ground Truth Used

  • The "ground truth" for the performance evaluation appears to be a combination of:
    • Objective Measurement Standards: The ANSI S3.22:2009 standard provides objective benchmarks for hearing aid characteristics.
    • Clinical Assessments/Patient Outcomes: For aspects like gain, speech understanding, sound quality, comfort, and ear health, the "ground truth" would be derived from clinical measurements and patient self-reports, compared against a baseline or a predicate device.
    • Expert Observations: Audiologists performed "usability evaluations" and clinical assessments which would form part of the "ground truth" for those specific metrics.

8. The Sample Size for the Training Set

  • Not applicable / Not specified. This device is a hardware hearing aid with an analog, digitally programmable circuit. While it uses digital programming, it is not an AI/ML algorithm that requires a "training set" in the modern sense of supervised learning for classification or prediction tasks. The "training" here refers to the engineering design, calibration, and parameter setting, not data-driven machine learning.

9. How the Ground Truth for the Training Set Was Established

  • Not applicable / Not specified. As explained above, there isn't a "training set" in the AI/ML context for this type of device. The "ground truth" for its design and programming would be based on established audiological science, hearing aid design principles, and engineering specifications.

§ 874.3300 Air-conduction hearing aid.

(a)
Identification. An air-conduction hearing aid is a wearable sound-amplifying device intended to compensate for impaired hearing that conducts sound to the ear through the air. An air-conduction hearing aid is subject to the requirements in § 800.30 or § 801.422 of this chapter, as applicable. The air-conduction hearing aid generic type excludes the group hearing aid or group auditory trainer, master hearing aid, and the tinnitus masker, regulated under §§ 874.3320, 874.3330, and 874.3400, respectively.(b)
Classification. Class I (general controls). This device is exempt from premarket notification procedures in subpart E of part 807 of this chapter, subject to the limitations in § 874.9.