LogoFDA 510(k) Search
Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K100762
    Device Name
    FLAT PANEL DETECTOR, MODEL FDR D-EVO
    Manufacturer
    FUJIFILM MEDICAL SYSTEMS U.S.A., INC.
    Date Cleared
    2010-07-15

    (120 days)

    Product Code
    MQB,COM,MOB
    Regulation Number
    892.1680
    Type
    Traditional
    Panel
    Radiology
    Reference & Predicate Devices
    K091435,K090318,K042023
    Why did this record match?
    Device Name :

    FLAT PANEL DETECTOR, MODEL FDR D-EVO

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

    The FDR D-EVO flat panel detector system is intended to capture for display radiographic images of human anatomy. It is intended for use in general projection radiographic applications wherever conventional film /screen or CR systems may be used. The FDR D-EVO is not intended for mammography, fluoroscopy, tomography, and angiography applications.

    Device Description

    FDR D-EVO flat panel detector is an indirect-conversion amorphous silicon (a-Si) portable flat panel detector (FPD) utilizing GOS (Gadolinium OxySulfide) as a schifillator. FDR D-EVO detector has Fuji's unique Irradiation Side Sampling system, delivering high image quality. D-EVO's 14x17" standard cassette size affords it the ability for use as a retrofit in any analog bucky and/or as an additional panel with fixed digital radiography systems, allowing a quick conversion to digital X-ray technology. The D-EVO features a detachable power source that enables easy positioning within the radiographic room within an upright bucky, table or as a free cassette. Data captured via operator console is sent electronically to the Fujifilm FDX Console to be displayed on the peraitor.

    AI/ML Overview

    Here's a breakdown of the acceptance criteria and study details for the Fujifilm FDR D-EVO Flat Panel Detector, based on the provided text:

    1. Acceptance Criteria and Reported Device Performance

    Acceptance CriteriaReported Device Performance
    All images must be deemed to be of diagnostic capability.All images were deemed to be of diagnostic capability.
    At least 90% of the scores from the D-EVO system are greater than or equal to -1.All readers satisfied the success criteria with 100% of the D-EVO image scores greater than or equal to -1. (Meaning 100% of scores were >= -1).

    2. Sample Size and Data Provenance

    • Test Set Sample Size: 30 image pairs (comparing FDR D-EVO with Fujifilm Carbon XL-2).
    • Data Provenance: Not explicitly stated, but the submission is from Fujifilm Medical Systems, USA, suggesting the study was likely conducted in the USA or with data relevant to the US market. The mention of "internal and international IEC testing requirements" suggests broader applicability, but the image quality reader study specifically compared against a cleared Fujifilm device, implying a focused internal validation. It is a retrospective study, as pre-existing images were used for comparison.

    3. Number and Qualifications of Experts for Ground Truth

    • Number of Experts: Three.
    • Qualifications of Experts: Board-certified radiologists.

    4. Adjudication Method

    The document does not explicitly describe an adjudication method. It states that all images were deemed of diagnostic capability by the reviewers, and all three radiologists individually met the success criteria with 100% of their scores. This suggests individual assessment rather than a formal adjudication process between the experts to establish a single ground truth score per image.

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

    While a reader study was performed with 3 board-certified radiologists, the document does not describe an MRMC comparative effectiveness study in terms of measuring reader improvement with AI vs. without AI assistance. The study directly compares the image quality of the FDR D-EVO system against a predicate device (Fujifilm Carbon XL-2) based on radiologists' scores. It doesn't involve AI assistance to human readers for a comparative effectiveness analysis.

    6. Standalone (Algorithm Only) Performance Study

    The document describes an "image quality reader study" where radiologists reviewed images. This is not a standalone (algorithm only) performance study. The performance is assessed by human interpretation of the images generated by the device.

    7. Type of Ground Truth Used

    The ground truth for the test set was established by expert consensus (or individual expert assessment, implicitly leading to a consensus that images were of diagnostic quality) of board-certified radiologists judging the "diagnostic capability" of the images and assigning scores.

    8. Sample Size for the Training Set

    The document does not provide information on the sample size used for the training set. The FDR D-EVO is a flat panel detector, not an AI algorithm in the context of this 510(k) submission. Therefore, it does not typically involve a "training set" in the same way an AI/ML algorithm would. The product description focuses on its physical and technical components and capabilities rather than learned predictive models.

    9. How Ground Truth for Training Set Was Established

    As mentioned above, the concept of a "training set" with established ground truth is not applicable to this device as described. The device is a hardware component for capturing radiographic images.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1