Search Results

The PeriCoach® OTC is a perineometer designed to treat stress, mild-moderate urge and mixed urinary incontinence in women, by strengthening of the pelvic floor muscles through exercise. This device provides biofeedback via smart phone technology.

The PeriCoach® OTC device consists of a rigid probe covered in a silicone sheath that is temporarily inserted into the vagina. Sensors located under the sheath measure the strength of contraction of the user's pelvic floor muscles. This information is then transmitted wirelessly to a smartphone application in order to provide real-time feedback to the user. It is a single patient, reusable device to be supplied over-the-counter.

The provided document describes the PeriCoach® OTC device. While it mentions the device meets acceptance criteria, it does not explicitly provide a table of acceptance criteria and reported device performance in the typical quantitative format for algorithmic performance metrics (e.g., sensitivity, specificity, AUC). Instead, the "acceptance criteria" are implied by the standards and bench tests it passed, and the "performance" is qualitative or relates to the device's physical and functional properties.

The document focuses on demonstrating substantial equivalence to a predicate device (PeriCoach® K143580) for a change from prescription use to over-the-counter (OTC) use, and software/firmware upgrades. Therefore, the "study" described is primarily focused on safety and equivalence, rather than a clinical trial proving specific diagnostic or treatment efficacy metrics typically found with AI/ML device submissions.

Here's an attempt to answer your questions based on the provided text, while acknowledging its limitations for an AI/ML context:

1. Table of Acceptance Criteria and Reported Device Performance

As noted, the document doesn't provide a typical quantitative performance table for an AI/ML diagnostic. Instead, the acceptance criteria are met by demonstrating compliance with standards and successful bench testing. The "performance" is stated qualitatively in terms of safety, usability, and functional equivalence.

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

• Test Set Description: The document refers to a "post-market user survey." This survey appears to be the primary "test set" for assessing suitability for OTC use and general user experience.
• Sample Size: The exact sample size for the post-market user survey is not specified in the provided text. It is only referred to as "a post market user survey."
• Data Provenance: The document does not specify the country of origin for the survey data. It states the submission is from Australia ("Analytica Pty Ltd," "St Leonards, NSW Australia"), which might imply the survey was conducted there, but this is not explicitly stated. The survey is characterized as "post market," indicating it's retrospective relative to the device's initial market release (as a prescription device).

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

• Not Applicable in this context. The "ground truth" for this device's performance isn't established by expert review of medical images or data requiring clinical expertise (like radiologists for AI). Instead, the "ground truth" for the post-market user survey would be the users' direct self-reported experience and feedback, and for the bench testing, it was engineering measurements against pre-defined specifications. Health professionals might have been involved in designing the survey or interpreting aggregated results, but they were not "experts establishing ground truth" in the diagnostic AI sense.

4. Adjudication Method for the Test Set

• Not Applicable. Since the "test set" is a user survey and bench testing, there's no diagnostic decision or interpretation of complex medical data requiring an adjudication method like 2+1 or 3+1.

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

• No such study was performed or is relevant. This device is a biofeedback perineometer, not an AI-powered diagnostic imaging tool that assists human readers. Therefore, an MRMC study comparing human readers with and without AI assistance is not applicable to this submission.

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

• Partially Applicable. The "sensor behavior test" and the overall bench testing could be considered "standalone" technical performance assessments of the device's core functionality (measuring pelvic floor muscle strength). However, the primary "performance" assessment mentioned for the OTC transition is the user survey, which inherently involves human interaction with the device. There isn't an "algorithm only" performance measured in the sense of an AI model making a diagnosis/prediction without human input.

7. The Type of Ground Truth Used

• Engineering Specifications / User Experience Data.
  • For bench testing: The ground truth was defined by engineering specifications and relevant industry standards (e.g., specific thresholds for drop tests, durability cycles, immersion resistance, and sensor accuracy within tolerances).
  • For the user survey: The "ground truth" was derived from self-reported user feedback and experience data regarding ease of use, understanding, and safety concerns. There was no clinical outcome data (e.g., pathology or long-term therapeutic outcomes) presented to substantiate the device's efficacy, as its purpose is to aid in exercise, not diagnose a condition. The claim is for "strengthening of the pelvic floor muscles through exercise" with biofeedback, implying efficacy via adherence to exercise, not a direct treatment within the device itself.

8. The Sample Size for the Training Set

• Not applicable / Not specified for an AI model. The document describes a medical device, not an AI/ML algorithm that undergoes a training phase with a specific dataset. The device's "performance" is based on its physical and software design validation, and user experience.

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

• Not applicable. As above, this is not an AI/ML algorithm with a traditional "training set." The device's design and functionality would be based on engineering principles and potentially prior clinical knowledge about pelvic floor muscle training, not a machine learning training process.

Ask a specific question about this device