Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K170524

    Validate with FDA (Live)

    Device Name
    Access AMH
    Date Cleared
    2017-11-13

    (264 days)

    Product Code
    Regulation Number
    862.1092
    Age Range
    All
    Reference & Predicate Devices
    Predicate For
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticPediatricDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Access AMH assay is a paramagnetic particle chemiluminescent immunoassay for the quantitative determination of anti-Müllerian hormone (AMH) levels in human serum and lithium heparin plasma using the Access Immunoassay Systems as an aid in the assessment of ovarian reserve in women presenting to fertility clinics. This system is intended to distinguish between women presenting with AFC (antral follicle count) values > 15 (high ovarian reserve) and women with AFC values ≤ 15 (normal or diminished ovarian reserve). The Access AMH is intended to be used in conjunction with other clinical and laboratory findings such as antral follicle count, before starting fertility therapy. The Access AMH is not intended to be used for monitoring of women undergoing controlled ovarian stimulation in an Assisted Reproduction Technology program.

    Device Description

    The Access AMH assay, Access AMH Calibrators, Access AMH QC, along with the Access 2 Immunoassay System analyzer comprise the Access Immunoassay System for the quantitative determination of anti-Müllerian hormone (AMH) levels in human serum and plasma.

    AI/ML Overview

    This document describes the Beckman Coulter Access AMH assay, a paramagnetic particle chemiluminescent immunoassay for the quantitative determination of anti-Müllerian hormone (AMH) levels. It is intended as an aid in assessing ovarian reserve in women presenting to fertility clinics to distinguish between women with AFC (antral follicle count) values > 15 (high ovarian reserve) and women with AFC values ≤ 15 (normal or diminished ovarian reserve). The document outlines various analytical and clinical studies to demonstrate the device's performance.

    Here's an analysis of the provided information, structured according to your request:

    1. Table of Acceptance Criteria and Reported Device Performance

    The document doesn't explicitly state "acceptance criteria" for each parameter, but it implies performance goals based on its design and comparisons. I will present the stated design goals/expectations as acceptance criteria and the reported performance.

    Acceptance Criteria (Design Goal / Expectation)Reported Device Performance
    Limit of Blank (LoB): ≤ 0.01 ng/mL (0.07 pmol/L)LoB: 0.0040 ng/mL (0.029 pmol/L)
    Limit of Detection (LoD): ≤ 0.02 ng/mL (0.14 pmol/L)LoD: 0.0098 ng/mL (0.07 pmol/L)
    Imprecision: Total imprecision ≤ 10% CV for concentrations ≥ 0.16 ng/mL; Total SD ≤ 0.032 ng/mL at concentrations < 0.16 ng/mLPrecision Summary (Access 2 - three lots combined): - Sample 1 (0.09 ng/mL): Total SD = 0.005, %CV = 5.5 - Sample 2 (2.61 ng/mL): Total SD = 0.075, %CV = 2.9 - Sample 3 (8.56 ng/mL): Total SD = 0.282, %CV = 3.3 - Sample 4 (17.16 ng/mL): Total SD = 0.502, %CV = 2.9
    Reproducibility: (Implied to be within acceptable clinical performance for multi-site use)Reproducibility Summary (Access 2, N=320): - Samples (0.48 to 16.16 ng/mL): Total Imprecision (SD and %CV) are provided, ranging from 2.2% to 3.2% CV. - QCs (1.00 to 15.22 ng/mL): Total Imprecision (SD and %CV) are provided, ranging from 2.3% to 3.2% CV.
    High-dose Hook Effect: No high-dose hook effectNo high-dose hook effect observed up to 1,000 ng/mL.
    Dilution Recovery: Average recovery of 100 ± 10% for 10-fold dilution across the assay range (0-24 ng/mL)Demonstrated to dilute recover across the range (0-24 ng/mL) with average recovery of 100 ± 10%.
    Limit of Quantitation (LoQ): ≤ 0.08 ng/mL (0.57 pmol/L)LoQ: 0.013 ng/mL (0.093 pmol/L)
    Linearity: Maximum deviation from linearity ≤ 5% for samples > 0.16 ng/mL, and ≤ 0.04 ng/mL for samples ≤ 0.16 ng/mLMaximum deviation from linearity of 4.8% for samples > 0.16 ng/mL. (Statement implies also met for samples ≤ 0.16 ng/mL)
    Analytical Specificity: No significant interference (exceeding 10% shift in dose) from listed substances at indicated concentrationsNo significant interference observed for tested substances at indicated concentrations.
    Cross-reactivity: No significant cross-reactivity (exceeding 5% cross reactivity) from listed substances at indicated concentrationsNo significant cross-reactivity observed for tested substances at indicated concentrations.
    Matrix Comparison: Slope equal to 1.00 ± 0.10 for different sample types (Lithium Heparin vs. Serum No Gel, Lithium Heparin vs. Serum Gel, Serum Gel vs. Serum No Gel)Slopes: 1.03 (Lithium Heparin vs. Serum No Gel), 1.05 (Lithium Heparin vs. Serum Gel), 0.99 (Serum Gel vs. Serum No Gel). All within ± 0.10 of 1.00.
    Method Comparison (vs. commercially available immunoassay): Strong correlationIntercept = 0.04 ng/mL, Slope = 1.03 (1.01 – 1.06), Correlation Coefficient (r) = 0.99
    Clinical Study (correlation of AMH to AFC): Correlation between AMH values and AFC results (r)r = 0.77, p < 0.0001
    Clinical Study (distinguishing AFC groups): Ability to distinguish women with AFC values > 15 vs. ≤ 15For AFC > 15 group, with AMH cutoff 1.77 ng/mL: Sensitivity = 88.9%, Specificity = 59.1%. Study supports distinguishing these groups.

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

    • Clinical Study Test Set:

      • Sample Size: 164 women.
      • Data Provenance: Prospective, multi-center, conducted at 13 geographically diverse sites in the US.
      • Inclusion Criteria: ≥ 21 and < 46 years of age (mean 34.9, range 23-45), in first cycle of ovarian stimulation for IVF or IVF/ICSI, both ovaries present, regular menstrual cycle, no confirmed PCOS. The mean BMI was 25.1 ± 5.3.
      • Measurements: AFC and AMH determined on days 2-4 of the same spontaneous menstrual cycle.
    • Analytical Performance Test Sets (Various):

      • Imprecision: N/A (samples were artificial/controls for statistical analysis of variability).
      • Reproducibility: N=320 (calculated from 4 instruments, 4 replicates, 2 runs/day, 10 days for 6 samples and 3 QC samples).
      • High-dose Hook Effect: N/A (spiked normal human serum).
      • Dilution Recovery: N/A (serum and plasma samples across the range).
      • Analytical Specificity / Cross-reactivity: N/A (serum samples spiked with interfering/cross-reacting substances).
      • Matrix Comparison: 57 matched sets of serum (gel and no gel) and plasma (lithium-heparin) samples.
      • Method Comparison: 121 values (samples) across the range of the assay.
      • Expected Reference Intervals: Prospectively procured from apparently healthy, non-pregnant females at five (5) geographically diverse sites. Sample sizes per age group: 120 (18-25), 131 (26-30), 120 (31-35), 123 (36-40), 126 (41-45). Subjects had regular menstrual cycles, both ovaries present, and proven fertility.

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

    The ground truth for the clinical study was Antral Follicle Count (AFC), determined by transvaginal ultrasonography. The document does not specify the number of experts (e.g., sonographers or radiologists) who performed or interpreted these ultrasounds, nor their specific qualifications. It only states an "AFC result was determined by transvaginal ultrasonography, which measured follicles of 2 to 10 mm in diameter."

    4. Adjudication method (e.g., 2+1, 3+1, none) for the test set

    Not applicable. The ground truth (AFC) was an objective measurement performed via ultrasonography, not a subjective interpretation requiring adjudication among human readers.

    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

    Not applicable. This device is an in-vitro diagnostic (IVD) immunoassay, not an AI-powered image analysis or diagnostic aid for human readers. It provides a quantitative AMH value.

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

    Yes, the device's performance is standalone. It's an automated immunoassay system that provides a quantitative AMH measurement directly. The results of the analytical and clinical studies demonstrate this standalone performance. The clinical study correlates the AMH values from the device with AFC, which is a separate clinical measurement, to demonstrate its utility as an "aid in the assessment." The device itself produces a direct, quantitative output.

    7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

    The primary ground truth for the clinical utility study was Antral Follicle Count (AFC), determined by transvaginal ultrasonography. This is a direct clinical measurement used to assess ovarian reserve.

    8. The sample size for the training set

    This is an immunoassay technology, not a machine learning algorithm that requires a "training set" in the conventional sense (e.g., for AI model development). The development and calibration of the assay would typically involve extensive analytical studies and optimization using various reagent lots, calibrators, and QC materials, rather than a distinct "training set" of patient data for model learning. The described studies are primarily validation and verification studies.

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

    Not applicable, as this is an immunoassay rather than a machine learning model requiring a "training set" with ground truth in the AI context. The assay itself measures AMH levels directly, and its analytical performance is established through rigorous laboratory testing and comparative studies against predicate devices and clinical parameters like AFC.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1