Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K231344
    Date Cleared
    2023-08-02

    (85 days)

    Product Code
    Regulation Number
    876.1500
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The ActivSight Intraoperative Imaging System (ActivSight) is intended to provide real-time endoscopic fluorescence and near infrared imaging. ActivSight enables surgeons to visually assess vessels, blood flow, and related tissue perfusion using fluorescence and near infrared imaging, and at least one of the major bile ducts (cystic duct, or common hepatic duct) using fluorescence, all during minimally invasive surgery.

    Fluorescence imaging of biliary ducts with ActivSight is intended for use with standard of care white light, and when indicated, intraoperative cholangiography. The device is not intended for stand-alone use for biliary duct visualization.

    Device Description

    The ActivSight Intraoperative Imaging System (ActivSight) is an accessory to existing commercial surgical laparoscope systems, including cameras and video processor units. ActivSight provides real-time endoscopic fluorescence and nearinfrared imaging. These imaging features allow surgeons to visually assess vessels, blood flow, and tissue perfusion (using fluorescence and near-infrared imaging), and to visually assess at least one of the major bile ducts (cystic duct, common bile duct, or common hepatic duct) using fluorescence. Fluorescence imaging is enabled through use of any commercially available Indocyanine Green (ICG). These visualization features are available for surgeons to use during minimally invasive surgery. ActivSight is intended to be used in a surgical environment.

    AI/ML Overview

    This report confirms that the ActivSight Intraoperative Imaging System (K231344) has been found substantially equivalent to its predicate device (K203550) by the FDA. This determination is primarily based on non-clinical testing related to a new sterilization method, as the device itself is identical in intended use and technology to the predicate. Therefore, detailed information regarding acceptance criteria and performance of an AI/human-in-the-loop system, as typically required for novel AI-powered medical devices, is largely not applicable to this specific submission.

    However, based on the limited information provided, we can infer some aspects and highlight what is not present given the nature of the submission. Since the provided text is a 510(k) clearance letter and summary for a modification (specifically, a new sterilization method) to an existing device, it does not contain the detailed clinical study data typically found in original submissions for AI-powered devices.

    Here's an analysis based on the context:

    1. A table of acceptance criteria and the reported device performance

    The provided documentation does not specify acceptance criteria and reported device performance related to diagnostic accuracy or clinical effectiveness, as this was a 510(k) submission for a modification (sterilization) rather than a de novo AI device.

    The non-clinical testing performed focused on:

    Acceptance Criteria (Implied)Reported Device Performance
    Sterility Assurance Level (SAL) of 10^-6Achieved SAL of 10^-6 with listed VHP cycles
    Functional reliability/performance after reprocessingNo impact on functional reliability or performance over multiple reprocessing cycles

    2. Sample size used for the test set and the data provenance

    For the sterilization efficacy testing:

    • Sample Size: Not explicitly stated, but typically involves a sufficient number of test articles (e.g., medical devices or biological indicators) to statistically validate a SAL of 10^-6.
    • Data Provenance: Not specified, but likely laboratory testing conducted under controlled conditions. This is not clinical data from patients.

    For functional testing after reprocessing:

    • Sample Size: Not explicitly stated, but usually involves testing multiple units through a predefined number of reprocessing cycles.
    • Data Provenance: Not specified, but likely laboratory testing.

    3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

    This is not applicable to this type of submission. Ground truth, in the context of diagnostic accuracy, would involve expert radiologists or pathologists interpreting images or samples. The testing here is for sterility and functional reliability, not diagnostic performance.

    4. Adjudication method for the test set

    This is not applicable as there are no expert interpretations requiring adjudication.

    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 MRMC study was performed or required for this 510(k) submission, as it relates to a modification of an existing imaging system's sterilization method, not its diagnostic or assistive AI capabilities. The device is an intraoperative imaging system that provides fluorescence and near-infrared imaging for surgeons, not an AI-powered diagnostic tool for human readers.

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

    Not applicable. The ActivSight system is described as an "accessory to existing commercial surgical laparoscope systems" and "enables surgeons to visually assess," indicating it's a visualization tool used by surgeons, not a standalone AI algorithm producing interpretations.

    7. The type of ground truth used

    For sterilization testing, the ground truth is established by microbiological methods (e.g., culturing biological indicators after sterilization to confirm kill rates). For functional testing, the ground truth is the device's operational specifications and performance metrics (e.g., image quality, light output, mechanical integrity).

    8. The sample size for the training set

    Not applicable. This device is not an AI algorithm that requires a training set. Even if the original predicate device (K203550) involved some form of image processing or enhancement, the provided document does not indicate that it is an AI/ML-based device that would undergo a training phase.

    9. How the ground truth for the training set was established

    Not applicable for the same reason as point 8.

    In summary:

    This 510(k) clearance is for a change to an already cleared medical device (specifically, its sterilization method). Therefore, the detailed performance data, acceptance criteria, and study designs typically associated with new AI/ML-driven diagnostic or treatment devices are not present in this document. The "study" mentioned here refers to the non-clinical validation of the new sterilization processes and confirmation of the device's continued functional reliability after these processes.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1