Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K222767
    Device Name
    PeekMed web (v1)
    Manufacturer
    Date Cleared
    2022-12-30

    (108 days)

    Product Code
    Regulation Number
    892.2050
    Reference & Predicate Devices
    Why did this record match?
    Device Name :

    PeekMed web (v1)

    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    PeekMed® web is a system designed to help healthcare professionals carry out pre-operative planning for several surgical procedures, based on their imported patients' imaging studies. Experience in usage and a clinical assessment is necessary for the proper use of the system in the revision and approval of the output of the planning.

    The multi-platform system works with a database of digital representations related to surgical materials supplied by their manufacturers.

    Device Description

    PeekMed® web is a system designed to help healthcare professionals carry out pre-operative planning for several surgical procedures, based on their imported patients' imaging studies. Experience in usage and a clinical assessment is necessary for proper use of the system in the revision and approval of the output of the planning.

    The multi-platform system works with a database of digital representations related to surgical materials supplied by their manufacturers.

    As PeekMed® web is capable of representing the medical images in a 2D or 3D environment, performing relevant measurements on those images and also capable of adding templates, it then can perform a total overview of the surgery. Being software it does not interact with any part of the body of the user and/or patient.

    AI/ML Overview

    The provided text describes PeekMed web (v1), a medical image management and processing system for pre-operative planning, and its substantial equivalence to a predicate device, PeekMed. The document focuses on regulatory compliance and the differences between the new device and its predicate.

    However, the furnished text does not contain the specific details required to answer all parts of your request, particularly regarding the acceptance criteria, the study design for proving the device meets it, and detailed performance metrics of the ML models. The document states that "ML models incorporated into PeekMed web were also trained, tested and validated for their performance," and that the "measuring function of the software was verified and validated... in order to assure the safety and correct performance of the device compared to the predicate." It also mentions fulfilling "previously defined accuracy and precision specifications." This implies that such studies were performed, but the results themselves and the specific criteria are not provided in this document.

    Therefore, the following response is based only on the information explicitly available in the provided text, and I will highlight where the requested information is not present.


    Device: PeekMed web (v1)

    Acceptance Criteria and Reported Device Performance

    The document broadly states that the device was validated to ensure it "fulfills the previously defined accuracy and precision specifications" for its measuring function and that the "ML models... were also trained, tested and validated for their performance." However, specific numerical acceptance criteria or reported performance metrics (e.g., sensitivity, specificity, or error margins for measurements) are NOT provided in the text.

    Acceptance Criteria (e.g., Performance Threshold for ML Models, Measurement Accuracy/Precision)Reported Device Performance
    Not explicitly stated in the provided text. The document indicates "previously defined accuracy and precision specifications" for the measuring function and general validation for ML model performance.Not explicitly stated in the provided text. The document states "it was confirmed that it fulfills the previously defined accuracy and precision specifications" for the measuring function and that ML models were "validated for their performance."

    Study Information

    1. Sample Size Used for the Test Set and Data Provenance:

      • Sample Size: Not specified in the provided text. The document mentions "All anatomical areas were tested, as well as other main areas of the software, such as the planning final report, and saved planning, ML models, among others." However, the number of cases/images used for testing is not detailed.
      • Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
    2. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:

      • Number of Experts: Not specified. The text mentions that "It is mandatory that the qualified user validates each individually landmatically positioned" and that "qualified users (trained surgeons) can perform activities related to the approval of clinical and critical information." This suggests expert involvement in ground truth establishment for validation/review, but the specific number and their qualifications for formal test set ground truth are not detailed.
      • Qualifications of Experts: The text refers to "qualified medical specialist (user)" and "trained surgeons." Specific experience levels (e.g., "10 years of experience") are not provided.
    3. Adjudication Method for the Test Set:

      • Not specified. The document states that "An automatic plan is always reviewed and validated by the qualified medical specialist" and that "It is mandatory that the qualified user validates each individually landmatically positioned." This describes the workflow of the device requiring user validation, but not a formal adjudication method (e.g., 2+1, 3+1) for establishing ground truth during the validation studies.
    4. 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, an MRMC comparative effectiveness study is not explicitly mentioned or detailed in the provided text. The document describes the automatic planning and landmarking features as "designed to improve and accelerate the user planning experience," but no results from MRMC studies showing an effect size of human reader improvement with AI assistance are provided.
    5. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done:

      • The document implies that standalone performance of the ML models was evaluated, as it states: "ML models incorporated into PeekMed web were also trained, tested and validated for their performance." However, the specific metrics and the definition of "standalone" in this context are not detailed. The device's use case heavily emphasizes human validation of the AI's output, suggesting standalone performance serves as a component of the overall system validation rather than a primary use case.
    6. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.):

      • The text suggests that for the automatic planning and landmarking features, the ground truth or the corrective mechanism involves validation by "qualified medical specialists" or "trained surgeons." For the measuring function, it was verified to "effectively and repeatedly match the real dimensions," implying comparison against a known or established standard/reference. There is no mention of pathology or outcomes data as ground truth.
    7. The Sample Size for the Training Set:

      • Not specified in the provided text. The document mentions that "For the development of these 2 ML models, it was verified that no pediatric images were used." This pertains to the exclusion criteria for the training data, but not its size.
    8. How the Ground Truth for the Training Set Was Established:

      • Not explicitly detailed in the provided text. It is implied that for the ML models, the training data would have had some form of annotated ground truth, likely established by experts, given the nature of pre-operative planning. However, the specific methodology (e.g., single expert, consensus) is not described.
    Ask a Question

    Ask a specific question about this device

    Page 1 of 1