Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K232976
    Device Name
    VUZE System
    Manufacturer
    Date Cleared
    2024-05-09

    (231 days)

    Product Code
    Regulation Number
    892.2050
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The VUZE System is intended to enable users to load pre-operative 3D images and planning data and register and overlay this data in real time with intra-operative 2D radiographic images of the same anatomy to support device guidance during interventional spinal procedures. The system also offers pre-operative surgical planning including implant sizing, entry location, and trajectory determination along with intra-operative guidance and tool trajectory / position confirmation.

    Device Description

    The VUZE System (the "System") enables users to load 3D images and planning data then register and overlay this data in real time with intra-operative 2D radiographic images of the same anatomy. The System supports device quidance during minimally invasive spinal surgery, including the stabilization of the spine by means of fixation. fixation coupled with fusion, vertebroplasty or kyphoplasty. Applicable vertebrae are within the range of S1 through T7.

    The System offers optional pre-operative surgical planning including implant sizing, entry location and trajectory determination, along with intra-operative guidance and tool trajectory/position confirmation by displaying a graphical representation of a tool tracked by intra-operative 2D images onto a patient's pre-operative 3D images.

    The System's main components include:

    • A workstation running the VUZE Planning and Procedure software (pre-installed)
    • A housing for the workstation, with a front door for user access as well as a back service door
    • A 32" touchscreen
    • A mouse
    • An isolation transformer
    • An internal video acquisition device (frame grabber)
    • A minimal-footprint, wheeled cart on which the above-listed items are placed.
    • DVI cable with galvanic isolator
    • An optional standalone planning station (planning software running on a commercial PC with identified specifications)
    • Optional C-arm orientation sensors (3), charger, and associated Bluetooth dongle (Note: The orientation sensor may also be referred to under an acronym of IMU)
    • Optional OTS foot pedal and associated Bluetooth dongle
    AI/ML Overview

    Acceptance Criteria and Study for the VUZE System (K232976)

    Based on the provided FDA 510(k) summary, the VUZE System is a medical imaging system for spinal interventions. The submission focuses on evaluating the substantial equivalence of a modified VUZE System to its predicate device (original VUZE System K210830). The performance data presented primarily assesses the quantitative accuracy of the system's core function: registering pre-operative 3D images with intra-operative 2D radiographic images to guide surgical tools.

    1. Table of Acceptance Criteria and Reported Device Performance

    The acceptance criteria are established as Root Mean Square (RMS) specification limits and quantile limits for various accuracy parameters. The reported device performance is based on testing at "Recommend Angles" and "Minimal Angle Difference" scenarios.

    ParameterAcceptance CriteriaReported Device Performance (Recommend Angles)StatusReported Device Performance (Minimal Angle Difference)Status
    RMS Specification Limits
    Direction error [deg]≤ 3°0.3094°Pass0.4537°Pass
    Tip deviation from GT trajectory [mm]≤ 2 mm0.2833 mmPass0.4107 mmPass
    Predicted tip deviation from GT trajectory [mm]≤ 2 mm0.3572 mmPass0.5199 mmPass
    Depth error [mm]≤ 4 mm0.8196 mmPass1.0194 mmPass
    Quantile Specification Limits (95% / 2.5% & 97.5%)
    Direction error [deg]< 5.7° (95th percentile)0.545° (95th percentile estimate)Pass0.851° (95th percentile estimate)Pass
    Tip deviation from GT trajectory [mm]< 2 mm (95th percentile)0.479 mm (95th percentile estimate)Pass0.798 mm (95th percentile estimate)Pass
    Predicted tip deviation from GT trajectory [mm]< 2 mm (95th percentile)0.615 mm (95th percentile estimate)Pass0.995 mm (95th percentile estimate)Pass
    Depth error [mm] Lower> -6.5 mm (2.5th percentile)-1.371 mm (2.5th percentile estimate)Pass-1.939 mm (2.5th percentile estimate)Pass
    Depth error [mm] Upper< 6.5 mm (97.5th percentile)1.920 mm (97.5th percentile estimate)Pass2.394 mm (97.5th percentile estimate)Pass

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

    The document does not explicitly state the numerical sample size (e.g., number of cases or images) used for the quantitative accuracy testing. It mentions "all combinations of X-Ray machines and CT/CBCT data" were used.

    Data Provenance: Not explicitly stated as "retrospective" or "prospective." The testing described is "Simulated Use / Quantitative Accuracy," suggesting it was conducted in a controlled environment as part of verification and validation, rather than directly from real-world patient data. The provenance of the CT/CBCT data and X-ray images used for these simulations (e.g., from a specific country or derived from anonymized patient data) is not specified.

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

    The document does not provide information on the number of experts or their qualifications for establishing the ground truth for the test set. The ground truth (GT) is referred to in "Tip deviation from GT trajectory" and "Predicted tip deviation from GT trajectory," implying a reference standard was used, but details about its establishment are absent. Given the type of accuracy metrics (e.g., tip deviation, direction error), it's highly likely a physical phantom or controlled simulation with precisely known geometries served as the ground truth.

    4. Adjudication Method for the Test Set

    The document does not mention an adjudication method (like 2+1, 3+1, or none) for the test set. This type of adjudication is typically used for image interpretation tasks involving multiple human readers, which is not the primary focus of the performance data presented here (which is quantitative accuracy of registration and tool trajectory).

    5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

    No, a multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly mentioned or described in the provided text. The performance data focuses on the standalone quantitative accuracy of the device's algorithmic functions rather than assessing the improvement of human readers with AI assistance.

    6. Standalone (Algorithm Only) Performance

    Yes, the performance data presented is a standalone (algorithm only) performance evaluation. The tables show the accuracy of the VUZE System in calculating direction error, tip deviation, predicted tip deviation, and depth error against a "GT trajectory" and specifications. This assesses the algorithmic output itself without specifying human interaction in the measurement of these specific performance metrics.

    7. Type of Ground Truth Used

    The type of ground truth used is implied to be a precisely defined geometric reference or simulated trajectory (e.g., from a phantom or highly accurate simulation model). The metrics "Tip deviation from GT trajectory" and "Depth error from GT" strongly suggest a known, exact reference point or path that the system's output is being compared against. It is not expert consensus, pathology, or outcomes data.

    8. Sample Size for the Training Set

    The document does not explicitly state the sample size for the training set. It describes "modifications to the CT to X-ray registration algorithm" and a "Deprecation of the previous ML algorithm... Same functionality is now done by a traditional algorithm." This implies that prior algorithms might have used training data, but the current submission's focus is on the modified algorithms and their direct performance. If the current critical algorithms are "traditional" (i.e., not machine learning based), then training data in the conventional sense might be less relevant for this submission, although development of such algorithms could involve extensive testing against diverse data.

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

    Since the document does not specify a training set sample size or details about a machine learning algorithm's training, it does not provide information on how the ground truth for a training set was established.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1