LogoFDA 510(k) Search
Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K192788
    Device Name
    ADVIA Centaur Cortisol (COR)
    Manufacturer
    Siemens Healthcare Diagnostics, Inc.
    Date Cleared
    2019-11-25

    (56 days)

    Product Code
    JFT
    Regulation Number
    862.1205
    Type
    Traditional
    Panel
    Clinical Chemistry
    Reference & Predicate Devices
    K142723
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    For in vitro diagnostic use in the quantitative determination of cortisol in serum, plasma (EDTA and lithium heparin), and urine using the ADVIA Centaur® XP system.

    Measurements of cortisol are used in the diagnosis and treatment of disorders of the adrenal gland.

    Device Description

    The ADVIA Centaur Cortisol (COR) assay is a competitive immunoassay using direct chemiluminescent technology. Results are determined using a calibration curve that is generated specifically on each instrument by a 2-point calibration and a master curve with the reagent bar code. The ADVIA Centaur Cortisol (COR) assay is intended for use on the ADVIA Centaur family of analyzers. The ADVIA Centaur Calibrator E is a set of 2 level calibrators for the assay. Siemens recommends the use of commercially available quality control materials with at least two levels (low and high).

    AI/ML Overview

    This document is focused on the ADVIA Centaur Cortisol (COR) assay, specifically the addition of plasma (EDTA and lithium heparin) as a sample claim. It seeks to demonstrate substantial equivalence to an existing device (K142723) which already had claims for serum and urine.

    Acceptance Criteria and Reported Device Performance

    The core of the study is to prove that the performance of the assay with the new plasma sample types is equivalent to its performance with serum (the established sample type). The primary acceptance criteria for this type of submission involve showing a strong correlation and minimal bias between the new sample type and the established one.

    Table of Acceptance Criteria and Reported Device Performance:

    Performance MetricAcceptance Criteria (Implicit from Industry Standards like CLSI EP09-A3)Reported Device Performance (ADVIA Centaur Cortisol (COR) with new plasma claims vs. Serum)
    Correlation Coefficient (r)Typically, a correlation coefficient (r) close to 1.00 (e.g., >0.975 or >0.98 is often considered good for method comparison studies) indicating a strong linear relationship between the new and established methods.Dipotassium EDTA Plasma vs. Serum: 1.00 Lithium-Heparin Plasma vs. Serum: 1.00
    Slope (from Deming Regression)A slope close to 1.00 (e.g., 0.95 to 1.05) indicating proportional agreement between the new and established methods.Dipotassium EDTA Plasma vs. Serum: 0.95 Lithium-Heparin Plasma vs. Serum: 0.96
    Intercept (from Deming Regression)An intercept close to 0 (e.g., within a predefined range that is considered clinically insignificant) indicating fixed bias between the new and established methods.Dipotassium EDTA Plasma vs. Serum: 0.24 µg/dL Lithium-Heparin Plasma vs. Serum: 0.56 µg/dL
    Bias from InterferentsBias should be within acceptable limits for clinical significance (e.g., < ±10% or clinically insignificant change). While not explicitly stated as an acceptance criterion in terms of a percentage, the results are presented to demonstrate minimal impact, aligning with good laboratory practice and CLSI EP07-ed3.Dipotassium EDTA (9.0 mg/mL): At 12.94 µg/dL: 0.5% bias At 50.39 µg/dL: -1.1% bias Heparin (75 U/mL): At 7.85 µg/dL: 2.9% bias At 46.50 µg/dL: -0.1% bias

    Study Details:

    1. Sample sizes used for the test set and data provenance:

      • Dipotassium EDTA Plasma vs. Serum: N = 83 samples
      • Lithium-Heparin Plasma vs. Serum: N = 99 samples
      • Data Provenance: The document does not explicitly state the country of origin or whether the data was retrospective or prospective. However, given that it's a 510(k) submission for commercial use, it's highly probable these were prospective studies conducted in a controlled environment, likely in the US or a country with similar regulatory standards for medical device development.

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

      • This submission is for an in vitro diagnostic (IVD) quantitative assay, not an AI/imaging device requiring expert interpretation for ground truth.
      • The "ground truth" for this type of device is established by the measurement itself using a reference method or the established method (serum measurement in this case). Therefore, there were no "experts" in the sense of radiologists or pathologists establishing subjective ground truth on the test set. The comparison is between different sample types on the same device.

    3. Adjudication method for the test set:

      • Not applicable. This is a quantitative assay comparison, not subject to human adjudication methods like consensus reading for imaging. The comparison is statistical (Deming regression, bias calculation) between objective numerical results.

    4. 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 is not an AI/imaging device. It's a chemistry assay for quantitative determination of cortisol.

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

      • Not applicable. This is a fully automated in vitro diagnostic assay. While the instrument performs the measurement "stand-alone" in terms of algorithm/chemistry, it's not an AI diagnostic algorithm in the context of imaging or clinical decision support where such a distinction is typically made. The performance presented is the "standalone" or instrument performance.

    6. The type of ground truth used:

      • The "ground truth" here is the measurement obtained from the established sample type (serum) using the same ADVIA Centaur Cortisol (COR) assay. The study aims to show equivalence of the new sample types (plasma) to this established serum measurement. This is a form of comparative measurement validation.

    7. The sample size for the training set:

      • Not applicable in the AI/machine learning sense. This is a traditional IVD assay based on competitive immunoassay and chemiluminescent technology. There is no "training set" in the context of deep learning models. The assay's chemical and optical parameters are inherently "tuned" during its development, but not using a machine learning training dataset.

    8. How the ground truth for the training set was established:

      • Not applicable due to the nature of the device (traditional IVD assay). The development and calibration of the assay involve rigorous internal validation using well-characterized samples and reference methods, but it's not a "training set" with "ground truth" established by human experts in the ML sense. The calibration curve is generated on each instrument via a 2-point calibration and a master curve. The master curve materials have 7 levels of cortisol, likely used for establishing the assay's quantitative response profile.
    Ask a Question

    Ask a specific question about this device

    Page 1 of 1