Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K240130
    Date Cleared
    2024-02-15

    (29 days)

    Product Code
    Regulation Number
    892.2050
    Reference & Predicate Devices
    Why did this record match?
    Device Name :

    Medical Monitor (21HQ613D)

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

    This Medical Monitor is indicated for use in displaying radiological images (including full-field digital mammography and digital breast tomosynthesis) for review, analysis, and diagnosis by trained medical practitioners.

    Device Description

    The Medical monitor is intended to provide high resolution color and grayscale medical imaging for PACS and Radiology system. This Medical Monitor is intended to be used by trained medical practitioners for displaying, reviewing, and analysis of medical images

    AI/ML Overview

    The LG Medical Monitor (Model 21HQ613D) is indicated for displaying radiological images, including full-field digital mammography and digital breast tomosynthesis, for review, analysis, and diagnosis by trained medical practitioners.

    Here's a breakdown of the acceptance criteria and the study conducted:

    1. Table of Acceptance Criteria and Reported Device Performance:

    The document states that "All display characteristics of 21HQ613D have met the pre-determined criteria." These criteria are implicitly defined by the successful "PASS" result for each measurement stated in the Non-Clinical Test Summary. The acceptance criteria are derived from the FDA guidance "Display Devices for Diagnostic Radiology".

    Measurement CategoryDescription (Acceptance Criteria)Reported Device Performance
    1. Spatial resolutionMeasurements of the transfer of information from the image data to the luminance fields at different spatial frequencies of interest, typically done by reporting the modulation transfer function. Non-isotropic resolution properties should be characterized properly by providing two-dimensional measurements or measurements along at least two representative axes.PASS
    2. Pixel defectsMeasurements (count, types (e.g., sub-pixel or entire pixel, always-on, always-off), and locations (map)) of pixel defects. This is typically provided as a tolerance limit. Pixel defects can interfere with the visibility of small details in medical images.PASS
    3. ArtifactsEvaluate for image artifacts such as ghosting and/or image sticking from displaying a fixed test pattern for a period of time.PASS
    4. Temporal responseMeasurements of the temporal behavior of the display in responding to changes in image values from frame to frame. Since these transitions are typically not symmetric, rise and fall time constants are needed to characterize the system. Slow displays can alter details and contrast of the image when large image stacks are browsed or in video, panning, and zooming modes. For mammography displays, rise and fall time constants at several (e.g., every 15 levels) grayscale intervals between 0 and 255 should be measured.PASS
    5. LuminanceMeasurements of the maximum and minimum luminance that the device outputs as used in the application under recommended conditions and the achievable values if the device is set to expand the range to the limit.PASS
    6. Conformance to a grayscale-to-luminance functionMeasurements of the mapping between image values and the luminance output following a target model response for 256 or more levels.PASS
    7. Luminance at 30° and 45° in diagonal, horizontal, and vertical directions at center and four cornersMeasurements of the luminance response at off-normal viewing related to the target model for the luminance response.PASS
    8. Luminance uniformity or Mura testMeasurements of the uniformity of the luminance across the display screen.PASS
    9. Stability of luminance and chromaticity response with temperature and time of operation (on-time)Measurements of the change in luminance and chromaticity response with temperature and use time.PASS
    10. Spatial noiseMeasurements of the spatial noise level as represented by the noise power spectrum using an appropriate ratio of camera and display pixels. Spatial noise and resolution affect the way images are presented to the viewer and can alter features that are relevant to the interpretation process of the physician or radiologist.PASS
    11. Reflection coefficientMeasurements of the reflection coefficients of the display device. Specular and diffuse reflection coefficients can be used as surrogates for the full bidirectional reflection distribution function.PASS
    12. Veiling glare or small-spot contrastMeasurements of the contrast obtained for small targets.PASS
    13. Color tracking (primary colors and color gamut)Chromaticity at different luminance levels of primary colors as indicated by the color coordinates in an appropriate units system (e.g., CIE u'v') and the color gamut enveloped by the primary colors.PASS
    14. Gray tracking (gray shades and white point)Chromaticity at different luminance levels of gray shades, including the white point, as indicated by the color coordinates in an appropriate units system.PASS

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

    The document describes a "Bench Test Performance Test" where "Physical Laboratory Test items suggested in the FDA guidance 'Display Devices for Diagnostic Radiology' were tested on 21HQ613D." This indicates that the testing was performed on the device itself (LG Medical Monitor, Model 21HQ613D). The sample size is not explicitly stated as a number of devices, but rather relates to the inherent characteristics of a single device under various measurement conditions. The data provenance is from laboratory testing of the physical device by LG Electronics Inc., South Korea. This would be considered prospective testing for the specific device model seeking clearance.

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

    This information is not provided in the document. The study described is a non-clinical bench test of the display device's performance characteristics, not a study involving human interpretation of images where ground truth would typically be established by expert readers.

    4. Adjudication Method for the Test Set:

    This information is not applicable to the described study. Adjudication methods are typically used in clinical studies involving multiple readers to resolve discrepancies in diagnoses or assessments. The reported study performed physical laboratory tests with specific objective measurements.

    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:

    A Multi Reader Multi Case (MRMC) comparative effectiveness study was not done. This submission is for a medical monitor, not an AI-powered diagnostic or assistive tool. The studies focused on the performance of the display hardware itself.

    6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was Done:

    A standalone algorithm-only performance study was not done in the context of diagnostic interpretations. The study focused on the performance of a medical display device, which is a "standalone" device in the sense that its performance characteristics are measured independently of a human user making a diagnosis. However, this is not an algorithm performing a diagnostic task.

    7. The Type of Ground Truth Used:

    The "ground truth" for the non-clinical bench tests was the objective physical measurements of the display's characteristics against pre-determined engineering and display quality standards outlined in the FDA guidance "Display Devices for Diagnostic Radiology." This is analogous to a reference standard for physical characteristics rather than a clinical diagnosis ground truth (e.g., pathology, outcomes data).

    8. The Sample Size for the Training Set:

    This information is not applicable. The device is a medical monitor, which is hardware for displaying images. It does not utilize a training set in the way a machine learning algorithm would. While its internal calibration tools (LG Calibration Studio Medical and DBI Calibration Feedback System) contain software, these are for maintaining display quality, not for image analysis requiring a training set for diagnostic tasks. The software for these tools was "designed and developed according to a software development process and was verified and validated" according to IEC 62304.

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

    This information is not applicable as there is no training set for a diagnostic algorithm.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1