Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K201141
    Device Name
    FIRMM
    Manufacturer
    Date Cleared
    2020-08-26

    (119 days)

    Product Code
    Regulation Number
    892.1000
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Nous FIRMM system is an accessory to an MRI scanner to calculate and display patient motion during a head scan. The motion results are derived from the MR image data as it is being acquired.

    MR images can be imported during the scan, and analyzed to detect patient motion in real-time. FIRMM enables the operator to become aware of patient motion during the scanning session and can be used to support scanning efficiency.

    The device is intended for prescription use only.

    Device Description

    The company has developed a software-based system that performs Framewise Integrated Real-time MRI Monitoring (FIRMM®). The FIRMM system continuously monitors the MR image data as it is being generated during the exam to detect small displacements (rotation or translation) of the anatomy along any axis during MRI acquisitions of time series datasets. It quantifies these movements and provides the MRI scanner operator with a plot of patient motion displayed as type of 'seismograph', as well as providing a figure of overall motion quality. This system allows for non-invasive, non-contact monitoring of patient motion during brain MRIs, and displays motion metrics in real-time to the scanner operator, with an optional ability to provide biofeedback to the patient.

    AI/ML Overview

    FIRMM Device Acceptance Criteria and Study Details

    1. Acceptance Criteria and Reported Device Performance

    ParameterAcceptance CriteriaReported Device Performance
    AccuracyExceeds specification (specific numerical criteria not provided in document)Accuracy exceeds the specification as demonstrated by processing synthetic data sets with known frame displacement (FD) values and comparing them to FIRMM's results.

    Note: The document only explicitly states accuracy criteria for the device's performance. Other performance aspects (e.g., compatibility, correlation with reference methods) are reported as having been studied and found satisfactory but without specific numerical acceptance values provided in this summary.

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

    The document mentions an "accuracy study" conducted using a "synthetic data set that included known frame displacement (FD) values." It also refers to a "4dfp comparison study" where "a wide variety of data sets from adult and infant subjects" were processed. However, specific sample sizes (number of cases or images) for the test sets in either study are not provided.

    The data provenance for the synthetic data set is by definition synthetic. For the "4dfp comparison study," the data is derived from "adult and infant subjects," implying retrospective clinical MR image data, though the country of origin is not specified.

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

    The document does not mention the use of experts to establish ground truth for the test sets.

    • For the accuracy study, ground truth (known FD values) was inherent to the synthetic data set.
    • For the 4dfp comparison study, ground truth was established by a "reference off-line method," not by human experts.

    4. Adjudication Method for the Test Set

    No adjudication method (e.g., 2+1, 3+1, none) for the test set is mentioned, as ground truth was established either synthetically or via a reference off-line method, not by human experts requiring adjudication.

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

    The document does not describe a multi-reader multi-case (MRMC) comparative effectiveness study involving human readers with and without AI assistance. The FIRMM system is described as an accessory to display patient motion to the operator and support scanning efficiency, rather than a diagnostic aid that directly impacts human reader interpretation of images. Consequently, there is no mention of an effect size for human reader improvement with AI assistance.

    6. Standalone Performance Study

    Yes, a standalone performance study was done for the algorithm. The "Accuracy" study directly assessed the device's ability to calculate and display patient motion by comparing its output (FD values) against known FD values from a synthetic dataset, without human intervention in the motion detection process. The "4dfp comparison study" also evaluated the algorithm's performance by comparing its results to a reference off-line method.

    7. Type of Ground Truth Used

    • For the accuracy study: Synthetic data with known frame displacement (FD) values.
    • For the 4dfp comparison study: A "reference off-line method." (Further details on this method are not provided in the summary).

    8. Sample Size for the Training Set

    The document does not specify the sample size for the training set. It describes the FIRMM system as a software-based system that uses MR image data for real-time motion detection, but it does not detail the machine learning methods (if any) or the training data used to develop the algorithm.

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

    The document does not provide information on how the ground truth for any potential training set was established. This information is typically relevant for machine learning models, and while FIRMM performs analytical tasks, the summary does not explicitly detail a training phase or the use of labeled training data with established ground truth.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1