LogoFDA 510(k) Search
Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K223841
    Device Name
    KBA3D
    Manufacturer
    S.M.A.I.O.
    Date Cleared
    2023-05-30

    (159 days)

    Product Code
    LLZ
    Regulation Number
    892.2050
    Type
    Traditional
    Panel
    Radiology
    Reference & Predicate Devices
    K210574,K213975,K180091,K141669
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The KBA3D v2.0.0 is intended for assisting healthcare professionals in viewing and measuring images as well as planning spine surgeries. The device allows surgeons and service providers to perform spine related measurements on images, and to plan surgical procedures. The device also includes tools for measuring anatomical components for design and placement of surgical implants. Clinical judgment and experience are required to properly use the software.

    The patient population targeted with the use of the KBA3D v2.0.0 software includes patients with mature skeletons requiring imaging measurements and planning of surgical procedures. The Bending algorithm is intended for patients older than 22 years old.

    Device Description

    The subject device KBA3D v2.0.0 is a SaaS software solution developed by S.M.A.I.O for the medical community; it is a second, independent version of the original KBA3D cleared by FDA in K213975. The current version of the KBA3D software is v.2.0.0. The user needs and requirements of the subject device were jointly defined by S.M.A.I.O and NuVasive. The software is intended to view images, perform spine related measurements, and plan surgical procedures such as osteotomies of the spine and templating of implants (screws, cages, rods).

    KBA3D V2.0.0 software can be used by health professionals (orthopedic surgeons, radiologists, neurosurgeons) and service providers (imaging technicians, clinical study technicians) who are trained in spine imaging and pathologies. The KBA3D V2.0.0 software is intended for patients requiring measurements and planning of surgical procedures. KBA3D v2.0.0 aims to achieve three objectives:

      1. From two perpendicular patient's standing x-rays including patient's spine and pelvis from the femoral heads to the cervical levels, provide 3D scaled representation of the femoral heads, sacral plate, and vertebral bodies. Provide related shape and positioning parameters measurements (disc/vertebra/height/angulation), main curvatures description and global balance assessment.
      1. Simulate potential effects of a spine surgery on spinopelvic alignment and provide related shape and positioning parameters calculation.
      1. Visualize scaled representation of implant range (pedicle screws, interbody cages, union rods) relative to spinopelvic representation (pre-op versus realigned) to establish possible implant selection scenarios.

    The software is not intended to predict the results of surgery as S.M.A.I.O does not provide tools to carry out planning. Therefore, regarding implant sizing, positioning, and correction impacts, accuracy levels provided by S.M.A.I.O are solely based on theoretical calculations that are not correlated to the output of surgery. KBA3D V2.0.0 provides scaled and simplified representations of the screws and cages relative to the patient's spine and pelvis.

    Bendini service (part of NuVasive Pulse System, K210574, reference device):

    • The first request from the software to the Bendini system enables the software to retrieve the most up-to-date benders/rods configuration file.
    • A second request from the software to the Bendini system enables the software to obtain specific bending instructions for the rod.

    Note: The Bendini Rod Bending algorithm is intended for patients older than 22 years old. The NuVasive Bendini Web App service is comprised of an API REST web service that facilities outside clients to utilize the NuVasive Bending algorithm. This algorithm is identical to the Bendini algorithm in K210574 for the NuVasive Pulse System and is hosted using the Microsoft Azure App Service. The KEOPS database manages connection to the software, patient data, and storage of data (X-rays and simulations).

    AI/ML Overview

    Here's a breakdown of the acceptance criteria and study information for the KBA3D v2.0.0 device, based on the provided text:

    1. Table of Acceptance Criteria and Reported Device Performance

    Feature/MeasurementAcceptance CriterionReported Device Performance
    3D Algorithm Reconstruction (Vertebral Body Dimensions from 2D Images vs. CT-Scan)Max deviation < 5mmMax average deviation was 4.3mm (most deviations < 3mm)
    Uncertainty Testing (Distance Deviation)< 2 pixels (px)Acceptable variations (specific value not given, but stated as "acceptable")
    Uncertainty Testing (Angular Deviation)< 2°Acceptable variations (specific value not given, but stated as "acceptable")
    Surgical Simulation (Spino-Pelvic Alignment, SSA error)Not explicitly stated, but implies high similarity to post-opHighest SSA error was 3.26%, back types obtained by simulation identical to post-op version
    Screw Positioning (Entry Point Deviation)< 0.5 mm< 10-12 mm (This seems to be a typo in the document as it's much larger than the acceptance criterion. Assuming it should be <0.5mm, or the criterion is misstated.)
    Screw Positioning (Angulation Deviation)< 0.5°< 0.3°
    Cage Positioning (Height Offset)< 0.5 mm< 0.4mm
    Cage Positioning (Angulation Deviation)< 0.5°< 10-12° (This also appears to be a typo in the document. Assuming it should be <0.5°, or the criterion is misstated.)
    Anatomical Parameters Measurements (compared to Excel calculations)No difference in calculationsNo difference in calculations
    Formative and Summative Studies (Safety & Usability)Intuitive use, successful performanceIntuitive use confirmed, users successfully performed measurements and 3D reconstruction that simulated surgery

    Important Note Regarding Typos: The reported performance for "Screw Positioning (Entry Point Deviation)" and "Cage Positioning (Angulation Deviation)" (< 10-12 mm and < 10-12° respectively) appears to conflict significantly with the acceptance criteria (< 0.5 mm and < 0.5°). It is highly probable these are typos in the provided document and likely intended to report performance within the acceptance criteria (e.g., < 0.01-0.02 mm or similar for the screws and cages). For accurate reporting, clarification from the original submission would be needed.

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

    • 3D algorithm reconstruction: Not explicitly stated, but refers to "Vertebral body dimensions... from CT-Scan."
    • Surgical Simulation: 12 patients (for comparison with post-op images).
    • Other tests (anatomical measurements, uncertainty, screw/cage positioning): Sample sizes are not explicitly provided in the document.
    • Data Provenance: The document does not specify the country of origin for the data or whether it was retrospective or prospective.

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

    • Surgical Simulation: "by a spine surgeon." The exact number (implied one or more), and detailed qualifications are not specified beyond "spine surgeon."
    • Other tests: The document does not explicitly state the number or qualifications of experts used for establishing ground truth for other tests.

    4. Adjudication Method

    The document does not explicitly describe an adjudication method (like 2+1, 3+1). For the surgical simulation, it mentions comparison "by a spine surgeon," implying assessment by an expert.

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

    No, a multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with AI assistance versus without AI assistance is not mentioned or described in the provided text. The device is referred to as "assisting healthcare professionals," but no study on the effectiveness of this assistance on human reader performance is detailed.

    6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

    Yes, a standalone performance assessment of the algorithm was conducted for various parameters:

    • Anatomical parameters measurements: Verified via mathematical calculations (Excel vs. KBA3D v2.0.0).
    • 3D algorithm reconstruction: Verified by comparison of Vertebral body dimensions from 3D reconstructions using the Balance Analyzer and 3D reconstructions from CT-Scan.
    • Uncertainty testing studies: Comparing original coordinates and "worst case variation" coordinates.
    • Performance of the algorithm to position the screws: Verified.
    • Performance of the algorithm to position the cage: Verified.

    7. Type of Ground Truth Used

    • Anatomical parameters measurements: Mathematical calculations (Excel).
    • 3D algorithm reconstruction: 3D reconstructions from CT-Scan.
    • Surgical Simulation: Post-treatment radiographs (post-op images) and assessment by a spine surgeon.
    • Screw & Cage Positioning: Theoretical calculations/references (implied by "performance of the algorithm to position the screws/cage was verified").

    8. Sample Size for the Training Set

    The document does not provide information on the sample size used for the training set.

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

    The document does not provide information on how the ground truth for the training set was established.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1