Search Results

Immunoassay for the in vitro quantitative determination of cortisol in human urine. The determination of cortisol is used for the recognition and treatment of functional disorders of the adrenal gland.

The electrochemiluminescence immunoassay "ECLIA" is intended for use on cobas e immunoassay analyzers.

The Elecsys Cortisol III immunoassay employs a competitive test principle using a cortisol-specific biotinylated monoclonal antibody and a cortisol-derivative labeled with a ruthenium complex. The Elecsys Cortisol III immunoassay is intended for the in vitro quantitative determination of cortisol in human urine. The determination of cortisol is used for the recognition and treatment of functional disorders of the adrenal gland on the cobas e immunoassay analyzers.

Results are determined via a calibration curve which is instrument-specifically generated by a two-point calibration and a master curve provided via the cobas link.

The Elecsys Cortisol III immunoassay reagent Rack Pack comprises the following:

M Streptavidin-coated microparticles (transparent cap), 1 bottle, 12.4 mL:
Streptavidin-coated microparticles 0.72 mg/mL; preservative.

R1 Anti-cortisol-Ab~biotin (gray cap), 1 bottle, 21.0 mL:
Biotinylated monoclonal anti-cortisol antibody (mouse) 18 ng/mL; danazol 20 μg/mL; MES buffer 100 mmol/L, pH 6.0; preservative.

R2 Cortisol-peptide~Ru(bpy) (black cap), 1 bottle, 21.0 mL:
Cortisol derivative (synthetic), labeled with ruthenium complex, 5 ng/mL; danazol 20 μg/mL; MES buffer 100 mmol/L, pH 6.0; preservative.

MES = 2-morpholino-ethane sulfonic acid

The provided 510(k) summary for the Elecsys Cortisol III device focuses primarily on non-clinical performance evaluations to demonstrate substantial equivalence to a predicate device. It does not describe a study to prove performance against specific acceptance criteria for diagnostic accuracy (e.g., sensitivity, specificity, or agreement with ground truth in a clinical context) with a test set of patient samples.

Here's an analysis of the available information:

1. Acceptance Criteria and Reported Device Performance

The document does not explicitly state "acceptance criteria" in the traditional sense for diagnostic performance metrics like sensitivity, specificity, or agreement against a clinical ground truth. Instead, it details performance specifications for various analytical aspects and states that these "met the predefined acceptance criteria." These are primarily related to the analytical performance of the assay itself.

2. Sample Size and Data Provenance for Test Set

• Precision (Repeatability & Intermediate Precision): Human urine samples (24-hour urine) and controls. Two replicates per run, two runs per day for 21 days for each of 4 human urine samples and 2 controls. (Total of $4 \text{ samples} \times 2 \text{ replicates/run} \times 2 \text{ runs/day} \times 21 \text{ days} = 336$ measurements for human urine, plus $2 \text{ controls} \times 2 \text{ replicates/run} \times 2 \text{ runs/day} \times 21 \text{ days} = 168$ measurements for controls. Or potentially 42 total runs for each sample/control).
• Analytical Sensitivity (LoB, LoD, LoQ): Not specified beyond "reagents and calibrators" likely being used.
• Linearity/Assay Reportable Range: Dilution series contained a minimum of 9 concentrations. Number of samples not explicitly stated but implies a set of samples specifically created to span the measuring range.
• HAMA: Not specified.
• Endogenous Interferences: Human urine samples (24-hour urine) were used. The number of samples is not explicitly stated.
• Analytical Specificity/Cross-Reactivity: Human urine (24-hour urine) samples. Specific numbers not provided beyond "samples were measured in the presence and absence of the potential cross-reactants."
• Exogenous Interferences – Drugs: In vitro tests performed on 12 commonly used pharmaceuticals and additional special drugs. This implies spiked samples rather than a "test set" of patient samples.
• Method Comparison: "Native 24 h urine samples" for comparison with the predicate device. The number of samples is not specified.
• Reference Range Study: Samples collected from an "apparently healthy population in the United States" across three study sites. The exact number of samples is not provided, but it's sufficient for establishing 2.5th and 97.5th percentiles (typically requires 120+ samples according to CLSI EP28-A3c).

Data Provenance: The document explicitly states "human urine samples (24-hour urine)" for most studies and for the reference range, "collected across three study sites... in the United States." This indicates prospective collection for the reference range study specifically for generating normal values applicable to the US population. For other analytical performance claims, the sample type (human urine) is generally mentioned, suggesting a similar provenance, likely for prospective testing within the manufacturer's lab or clinical sites.

3. Number of Experts and Qualifications for Ground Truth

Not applicable for the Elecsys Cortisol III. This is an in vitro diagnostic device (IVD) that quantitatively measures a biomarker (cortisol). The "ground truth" for such devices is typically established through recognized analytical standards, reference methods, and comparison to a legally marketed predicate device, rather than expert consensus on diagnostic images or clinical assessments. The closest to "ground truth" in this context would be the accuracy against a gold standard analytical method or purified cortisol standards. These details are not provided but are implicit in the validation that relies on CLSI guidelines.

4. Adjudication Method for the Test Set

Not applicable. As this is a quantitative IVD for a biomarker, diagnostic classification and adjudication by experts are not relevant to the described analytical studies. The performance is assessed by comparison to expected analytical results or a predicate device.

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

Not applicable. MRMC studies are typically for imaging devices or software that assist human readers in making a diagnosis. The Elecsys Cortisol III is an automated in vitro diagnostic immunoassay for quantitative measurement of cortisol in urine. It does not involve human readers interpreting cases with or without AI assistance.

6. Standalone Performance Study

Yes, the entire submission describes standalone performance. The Elecsys Cortisol III is an immunoassay designed to operate on cobas e immunoassay analyzers. All the performance data (precision, sensitivity, linearity, interference, cross-reactivity, method comparison) are generated directly from the device's measurement of cortisol in urine samples. The device itself performs the quantitative determination without human-in-the-loop interpretation impacting the measurement results. The method comparison study directly compares its quantitative output to the predicate device's quantitative output.

7. Type of Ground Truth Used

For an IVD like Elecsys Cortisol III, the "ground truth" for the test set is established by:

• Reference standards/Calibrators: For analytical sensitivity (LoB, LoD, LoQ) and linearity studies, known concentrations of cortisol (or materials traceable to them) are used.
• Predicate device comparison: For method comparison, the results from the Elecsys Cortisol III are compared to those obtained from the legally marketed ARCHITECT Cortisol (K062204), which serves as the established "truth" or benchmark for demonstrating substantial equivalence.
• Spiked samples/characterized samples: For interference and cross-reactivity studies, samples with known concentrations of interferents or cross-reactants are used to determine the device's accuracy in their presence.
• Clinically characterized healthy population samples: For the reference range study, samples from healthy individuals are used to establish normal ranges, though this isn't a "ground truth" for diagnostic accuracy.

8. Sample Size for the Training Set

The document does not mention "training set" in the context of an AI/ML algorithm. This device is an immunoassay, which relies on chemical reactions and optical detection, not an AI/ML model that requires a training set. The term "training set" is therefore not applicable here.

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

Not applicable. As noted above, there is no AI/ML training set for this immunoassay device.

Ask a specific question about this device

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.

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).

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:

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 specific question about this device

Immunoassay for the in vitro quantitative determination of cortisol in human serum, and plasma. The determination of cortisol is used for the recognition and treatment of functional disorders of the adrenal gland. The electrochemiluminescence immunoassay "ECLIA" is intended for use on Elecsys and cobas e immunoassay analyzers.

Cortisol II CalSet is used for calibrating the quantitative Elecsys Cortisol II assay on the Elecsys and cobas e immunoassay analyzers.

The Elecsys Cortisol II assay makes use of a competition test principle using a monoclonal antibody which is specifically directed against cortisol. Endogenous cortisol which has been liberated from binding proteins with danazol competes with exogenous cortisol derivative in the test which has been labeled with ruthenium complex for the binding sites on the biotinylated antibody.

Results are determined via a calibration curve which is instrument specifically generated by 2point calibration and a master curve provided via reagent barcode.

The reagent working solutions include:

• rackpack (kit placed on instrument) .
  • Streptavidin coated microparticles, ş
  • Reagent 1 (Anti-cortisol-Ab~biotin) and ş
  • Reagent 2 (Cortisol-peptide~Ru(bpy)2+3). ş

The Cortisol II CalSet is a lyophilized human serum with added cortisol in two concentration ranges.

The CalSet includes:

• Cal 1 (approximately 12.5 nmol/L cortisol in a human serum matrix) .
• Cal 2 (approximately 1000 nmol/L cortisol in a human serum matrix) .

The provided document describes the Elecsys Cortisol II immunoassay and its associated calibrator, Cortisol II CalSet. It details various non-clinical performance evaluations and one clinical performance evaluation.

Here's an analysis of the acceptance criteria and the studies that prove the device meets them:

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

The document does not explicitly state acceptance criteria in a dedicated table for each technical performance study. Instead, it presents the results of these studies and implies that these results demonstrate the device's performance is acceptable and supports substantial equivalence to the predicate device. For analytical specificity and method comparison, the document includes tables comparing some key characteristics with the predicate device. For other studies like precision, linearity, and sensitivity, it reports the performance without directly listing a pre-defined acceptance criterion.

However, I can extract the reported performance directly from the document for key metrics.

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

• Precision (Human Serum)
  • Test Set Sample Size: 5 human serum samples (native, diluted, spiked) and 2 controls (PC Universal).
  • Data Provenance: Not explicitly stated, but implies laboratory testing internal to the manufacturer (Roche Diagnostics).
• Limit of Blank (LoB)
  • Test Set Sample Size: 5 blank samples. A total of n = 60 LoB measurements were made.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Limit of Detection (LoD)
  • Test Set Sample Size: 5 low-level human serum samples (diluted). A total of n = 60 LoD measurements were made.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Limit of Quantitation (LoQ)
  • Test Set Sample Size: 9 human serum samples.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Linearity (Serum and Plasma)
  • Test Set Sample Size: 19 concentrations (thereof 17 dilutions) prepared from a high analyte serum/plasma sample. Each assayed in 3-fold determination within a single run.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Analytical Specificity (Serum/Plasma)
  • Test Set Sample Size: Two human serum sample pools.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Endogenous Interferences
  • Test Set Sample Size: Three human serum samples (low, mid, high cortisol concentrations) for each interfering substance.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• HASA Effect
  • Test Set Sample Size: Two serum samples with cortisol concentrations of 173 and 789 nmol/L. Aliquots spiked with DASA and diluted in 10% increments.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Exogenous Interferences - Drugs
  • Test Set Sample Size: Two human serum samples (native, diluted, spiked) for 16 pharmaceutical compounds; each tested in 3-fold determination.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Exogenous Interferences - Anticoagulants
  • Test Set Sample Size: Minimum of 48 serum/plasma pairs per sample material (Serum, Li-Heparin, K2-EDTA-, K3-EDTA-plasma primary tubes and Li-Heparin Plasma Gel Separation Tubes). Each tested in duplicate.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Method Comparison 1: Elecsys Cortisol II vs. LC-MS (Reference Method)
  • Test Set Sample Size: 208 human serum samples (all native single donors).
  • Data Provenance: UZ Gent, Lab voor Klinische Biologie (Gent, Belgium); implies prospective collection for the study or a well-characterized archive.
• Method Comparison 2: Elecsys Cortisol II vs. Cortisol I (K070788)
  • Test Set Sample Size: 536 human serum samples (all native single donors).
  • Data Provenance: UZ Gent, LMU Großhadern, Uni Klinik Leipzig and Labor Limbach; implies prospective collection for the study or a well-characterized archive from multiple European sites.
• Reagent Stability (Onboard and After First Opening)
  • Test Set Sample Size: 5 human serum samples (native, diluted, spiked) and 2 controls.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Reagent Stability (Real-time, ongoing)
  • Test Set Sample Size: PreciControl Universal (Level 1 and 2) and human serum samples. Data for time-points at 0, 13, 16, 19 months (MP lot) and 0, 7, 16, 19 months (P2 and P3 lot)
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Sample Stability (2-8°C, Room Temperature, -15 to -25°C)
  • Test Set Sample Size: Twelve human serum, Li-Heparin, K2-EDTA, and K3-EDTA plasma samples (all single donors, native, spiked, diluted) for each study.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Calibration Stability (Lot and On-board)
  • Test Set Sample Size: Five human serum samples (native, diluted, spiked) and two control samples.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Calibrator studies (Reconstitution, Stability)
  • Test Set Sample Size: Not explicitly stated beyond "two sets of Cortisol II CalSet" for reconstitution, and "on-test and reference materials" for stability. PreciControl Universal used to assess CalSet stability.
  • Data Provenance: Not explicitly stated, implies internal laboratory testing.
• Clinical Performance Evaluation (Reference Range)
  • Test Set Sample Size: 296 individuals for 6-10 am sample collection, 300 individuals for 4-8 pm sample collection.
  • Data Provenance: Three sites in the United States (one site in St Louis, Missouri for evaluation). These are prospective clinical samples.
• Training Set Sample Size: The document does not specify a separate "training set" for the Elecsys Cortisol II device. Immunoassays are not typically "trained" in the machine-learning sense with a distinct training dataset. Instead, they are developed, optimized, and then validated with performance studies. The "value assignment" for calibrators, and the master curves are part of the setup, but not a "training set" in the context of AI.

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

This information is not applicable or explicitly stated in the document for this type of medical device (an in-vitro diagnostic immunoassay). For immunoassays, ground truth (or reference values) is typically established through:

• Reference methods: Like LC-MS (Liquid Chromatography-Mass Spectrometry) as used in Method Comparison 1. These are highly accurate analytical techniques, not dependent on expert interpretation.
• Spiking studies/Dilution linearity: Known concentrations are added to samples, which serves as the "ground truth" for evaluating recovery and linearity.
• Clinical context: For reference ranges, the "truth" is derived statistically from a population of healthy individuals.
• Previous assay values: For comparative studies, the predicate device's results are used as a comparison point.

There are no mentions of human experts establishing "ground truth" in terms of interpreting clinical findings or images.

4. Adjudication method for the test set

Not applicable. This is not a device where human interpretation or adjudication is used to establish "ground truth" for the test results. The device measures a biomarker concentration.

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 document describes an immunoassay and calibrator, not an AI device that assists human readers in interpreting cases. There is no mention of human readers or AI assistance.

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

This refers to an immunoassay, which is by nature a standalone analytical system. The Elecsys Cortisol II assay determines the quantitative concentration of cortisol in human serum and plasma directly. There is no human-in-the-loop performance component in the measurement process itself, although clinical interpretation of the results by a healthcare professional is expected.

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

The types of "ground truth" or reference standards used in the studies include:

• Reference Method LC-MS: For Method Comparison 1, the Elecsys Cortisol II was compared against LC-MS, which serves as a highly accurate reference method for cortisol quantification.
• Predicate Device: For Method Comparison 2, the Elecsys Cortisol II was compared against the predicate Elecsys Cortisol (K070788) assay.
• Known Spiked/Diluted Concentrations: For studies like linearity, analytical specificity, and interference, known amounts of analyte or interfering substances were added to samples, establishing controlled "ground truth" for evaluation.
• Statistical Normative Data: For establishing the reference ranges, the results from a large cohort of self-reported healthy individuals were used to statistically define normal limits.
• Assigned Values: For calibrator and control stability, "assigned values" from independent measurements (e.g., using multiple analyzers and lots) were used as the reference.

8. The sample size for the training set

The document does not specify a "training set" in the context of machine learning for this immunoassay. Immunoassays typically involve development and optimization phases for reagent formulation, antibody selection, and instrument parameters, which could be seen as analogous to "training" in a broad sense, but not with a distinct training data set size as in AI/ML validation studies. Instead, the assay relies on a master curve (provided via reagent barcode) and 2-point calibration.

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

As there is no distinct "training set" in the AI/ML sense, this question is not directly applicable. However, for the reference values that define the assay's performance and calibration, the following methods are mentioned:

• Master curve: Provided via reagent barcode for quantitative determination.
• Calibration: Results are determined via a calibration curve that is instrument specifically generated by 2-point calibration.
• Traceability: The Elecsys Cortisol II assay has been standardized against the IRMM (Institute for Reference Materials and Measurements)/IFCC 451 panel (ID GC/MS, isotope dilution-gas chromatography/mass spectrometry), which serves as a highly accurate reference.
• Calibrator Value Assignment: For the Cortisol II CalSet, target values are chosen for the best fit with the Master Calibration Curve and Rodbard curve parameters. Calibrators are run in duplicate on multiple analyzers (at least 3 cobas e 411 and at least 3 cobas e 601/cobas e 602/MODULAR ANALYTICS E170 analyzers) with all available reagent lots. The assigned value is the mean over at least six runs on at least three analyzers.

Ask a specific question about this device

The ADVIA Centaur® Cortisol (COR) Assay is for in vitro diagnostic use in the quantitative determination of cortisol in serum or urine using the ADVIA Centaur XP system. Measurements of cortisol are used in the diagnosis and treatment of disorders of the adrenal gland.

The ADVIA Centaur Cortisol 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 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).

Here's a breakdown of the acceptance criteria and the study details for the ADVIA Centaur® Cortisol (COR) Assay, based on the provided FDA 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Precision: 80 replicates per sample type (controls, serum, urine pools) over 20 days.
• Linearity/Assay Range: Patient serum and urine samples (number not explicitly stated but "equally spaced dilutions across the assay range" were used and assayed in triplicate).
• Analytical Detection Limits (LoB, LoD, LoQ):
  • LoB: 6 blank samples, 5 replicates/day over 3 days (n=90).
  • LoD: 5 low cortisol serum samples, 5 replicates/day over 3 days (n=75).
  • LoQ: 6 samples with GCMS assigned doses, 5 replicates/day over 3 days (n=60).
• Analytical Specificity (Interference): 2 sample pools per interference (serum and urine) for each interferent, run in triplicate.
• Analytical Specificity (Cross-reactivity): 2 human serum sample pools per cross-reactant, run in triplicate.
• Expected Values (Reference Intervals):
  • New Serum Intervals: 252 serum samples (127 AM, 125 PM) from apparently healthy individuals.
  • Verified Urine Intervals: 20 24-hour direct urine specimens and 20 24-hour extracted urine specimens.
• Method Comparison:
  • 243 serum samples
  • 98 24-hour direct urine samples
  • 111 24-hour extracted urine samples
• Dilution Recovery: 5 human serum samples.

Data Provenance: The document does not explicitly state the country of origin for the data (e.g., patient samples for reference intervals, linearity, method comparison). However, it is an in vitro diagnostic device for global use, and such studies are typically multicenter or at least conducted with diverse populations relevant to the intended market. Given Siemens' global presence, it's likely a well-controlled study, but specific geographical details are not provided. The studies appear to be prospective as they were specifically designed and executed to evaluate the performance of the modified device against defined protocols.

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

There were no human "experts" establishing ground truth in the traditional sense (e.g., radiologists interpreting images). This device is a quantitative in vitro diagnostic assay. The ground truth for its performance characteristics is established through:

• Reference Methods: Such as GC-MS for traceability and potentially for assigning values to samples for linearity or detection limit studies.
• Calibrators and Controls: Professionally manufactured and value-assigned reagents.
• Statistical Analysis: CLSI guidelines (EP05-A2, EP06-A, EP17-A2, EP28-A3c, EP07-A2) dictate the statistical methods to define acceptance ranges and assess performance.
• Predicate Device: For method comparison, the predicate device acts as a reference standard.

Therefore, the "ground truth" is derived from established analytical methodologies and accepted clinical laboratory standards, rather than expert consensus on individual cases.

4. Adjudication Method for the Test Set

Not applicable. As a quantitative in vitro diagnostic assay, there is no "adjudication" of results in the way there is for image interpretation by clinicians. Results are numerical measurements subjected to statistical analysis and comparison against predefined performance criteria or reference methods.

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 is an in vitro diagnostic assay, not an AI-powered diagnostic imaging device that assists human readers.

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

Yes, the studies presented are all standalone performance evaluations of the ADVIA Centaur® Cortisol (COR) Assay system. The device itself performs the quantitative determination of cortisol in serum or urine. There is no "human-in-the-loop" performance component described in the context of the assay's function.

7. The Type of Ground Truth Used

The primary types of "ground truth" used are:

• Reference Method Traceability: Especially evident in the standardization of internal standards to Gas Chromatography-Mass Spectrometry (GC-MS).
• Assigned Values: For calibrators, controls, and Master Curve Materials.
• Statistical Acceptance Criteria: Defined by CLSI (Clinical and Laboratory Standards Institute) guidelines, which represent a consensus on best practices and acceptable performance limits in clinical laboratories.
• Predicate Device Measurements: Used as a comparative gold standard in the method comparison study.

8. The Sample Size for the Training Set

The document does not explicitly delineate a "training set" in the context of an AI/machine learning model. This device is a competitive immunoassay based on established chemical and immunological principles, not a machine learning algorithm that is "trained" on data in the conventional sense.

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

Not applicable. As mentioned above, there isn't a "training set" for an assay of this type. The "ground truth" for the method's development and validation (e.g., antibody specificity, reagent concentrations, instrument parameters) would have been established through extensive research, development, and internal testing by Siemens Healthcare Diagnostics. The studies presented here are validation studies to demonstrate the final product's performance and substantial equivalence.

Ask a specific question about this device

ARCHITECT® Cortisol is a chemiluminescent microparticle immunoassay (CMIA) for the quantitative determination of cortisol in human serum, plasma or urine on the ARCHITECT i System. The ARCHITECT Cortisol assay is intended for use as an aid in the diagnosis and treatment of adrenal disorders.

The ARCHITECT Cortisol Calibrators are for the calibration of the ARCHITECT i System when used for the quantitative determination of cortisol in human serum, plasma or urine.

The ARCHITECT Cortisol assay is a delayed one-step immunoassay for the quantitative determination of cortisol in human serum, plasma or urine using CMIA technology with flexible assay protocols, referred to as Chemiflex®.

Here's a breakdown of the acceptance criteria and study details for the ARCHITECT Cortisol assay:

1. Table of Acceptance Criteria and Reported Device Performance:

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

• Linearity: The study used a "regression analysis plot" and involved constructing "r regression and regression standard error (Reg SE)" for "each pool." The number of unique samples or data points for each dilution is not explicitly stated in the table, but it shows results for dilution range from 0% to 100% for two serum pools (65 ug/dL and 8 ug/dL). The guidance used was NCCLS guideline EP6-A. Provenance is not specified (e.g., country, retrospective/prospective).
• Accuracy: Cortisol was spiked into human serum from 3 donors and human urine from 3 donors. The samples were analyzed in triplicate. Provenance is not specified.
• Sensitivity:
  • LoB and LoD: 60 blank samples and 120 low-level samples were used to determine LoB and LoD.
  • Functional Sensitivity: Not explicitly stated, but based on CLSI guideline NCCLS EP17-A.
    Provenace of samples not specified.
• Method Comparison: Patients samples were used (serum and urine). The number of samples is not provided. The study was conducted according to CLSI Guideline NCCLS EP9. Provenance not specified.
• Precision:
  • 80 replicates (n=80) for each of the three MCC (Multi-constituent Control) levels and three serum panel levels on both the i2000 and i2000SR instruments.
  • 80 replicates (n=80) for each of the four urine panel levels on both the i2000 and i2000SR instruments.
    The study was performed using NCCLS guideline EP5-A2. Provenance not specified.
• Interferences (Endogenous Substances):
  • Serum: 3 replicates (N=3) for each interferent concentration and Cortisol target concentration.
  • Urine: Data represents a comparison of "Unaltered Urine Control" vs. "Mock spike Control" and then individual interferents. Number of replicates not explicitly stated, but implies multiple measurements.
    The study was conducted using CLSI Guideline NCCLS EP7-A2. Provenance not specified.
• Interferences (HAMA & RF):
  • HAMA: 10 samples (HAMA-1, HAMA-2 and 8 other samples) were tested.
  • Rheumatoid Factor (RF): 10 positive RF patient samples were assayed.
    Provenance not specified.
• Specificity: Cortisol was spiked into cortisol-free human serum. Cross-reactant stock concentrates were prepared and spiked into aliquots of the 12 ug/dL cortisol serum. The total number of unique samples/aliquots is not precisely stated but involves a comprehensive list of 37 cross-reactants. Provenance not specified.
• On-Board Stability: Details on the sample size used for stability studies (number of samples, replicates, etc.) are not provided in the summary. Provenance not specified.

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

The device is an in vitro diagnostic (IVD) assay for quantitative determination of cortisol. The "ground truth" for such devices is typically established through a reference method or known concentrations of analytes, not through human expert interpretation of images or clinical cases. Therefore, the concept of "experts" as in "radiologists with 10 years of experience" is not applicable here.

• For studies like linearity, accuracy (recovery), sensitivity, and precision, the ground truth is based on:
  • Known concentrations: For spiked samples and controls.
  • Reference method values: For method comparison (in this case, the predicate device Abbott AxSYM Cortisol assay).
• No information is provided about clinical expert adjudication or establishment of ground truth in the context of clinical expert review.

4. Adjudication Method for the Test Set:

Not applicable. As this is an IVD assay measuring an analyte concentration, the results are quantitative and compared against expected values, known concentrations, or results from a predicate device. There is no mention of a human expert adjudication process (e.g., 2+1, 3+1) because that process is typically associated with subjective interpretation tasks (like image review) where there can be disagreement among experts.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size:

No. An MRMC study is not relevant for this type of IVD device. MRMC studies are typically used to evaluate the impact of AI on human reader performance, for instance, in diagnostic imaging where human readers interpret cases. This submission is for a standalone laboratory assay that quantifies a biomarker.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done:

Yes, the studies presented (linearity, accuracy, sensitivity, precision, interferences, specificity, stability) all represent the standalone performance of the ARCHITECT Cortisol assay without human-in-the-loop assistance in the interpretation of the assay result itself. The assay directly outputs a quantitative value. The human involvement is in performing the test and reviewing the quantitative result, but not in interpreting subjective data for diagnostical decision-making.

7. The Type of Ground Truth Used:

The ground truth used for these studies includes:

• Known (prepared) concentrations: For linearity panels, spiked accuracy samples (serum and urine), and sensitivity studies (blank and low-level samples).
• Reference method/predicate device comparison: For method comparison, the Abbott AxSYM Cortisol assay served as the comparative "truth" or reference method to demonstrate substantial equivalence.
• Defined control material values: For precision studies (MCC samples).

8. The Sample Size for the Training Set:

This information is not applicable to this type of traditional in vitro diagnostic assay. These assays are developed through chemical and biological formulation and optimization, not through machine learning models that require distinct training sets. The development process involves extensive experimentation and validation, but not in the "training set" sense of AI/ML.

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

This information is not applicable for the same reason as above. There is no "training set" in the context of machine learning for this device. The development and optimization of the assay's reagents and protocols are based on established analytical chemistry principles and performance characteristics, validated against known standards and reference materials.

Ask a specific question about this device

The AxSYM® Cortisol assay is a Fluorescence Polarization Immunoassay (FPIA) for the quantitative measurement of Cortisol in human serum, plasma or urine on the AxSYM® System to aid in the diagnosis and treatment of adrenal disorders.

The AxSYM Cortisol assay utilizes Fluorescence Polarization Immunoassay (FPIA) technology. The AxSYM Cortisol Reagents and sample are pipetted in the following sequence:

• Sample and all AxSYM Cortisol Reagents required for one test are pipetted by the sampling probe into various positions of a Reaction Vessel (RV).
• Sample, the Cortisol Antiserum (antibody), pretreatment solution and Solution 4 (Line Diluent) are pipetted into one well of the RV.
• Additional pretreatment solution 4 (Line Diluent) are pipetted to the cuvette of the RV.
• The RV is immediately transferred into the Processing Center. Further pipetting is done in the Processing Center by the Processing Probe.
  In the processing center, an aliquot of the predilution mixture and Solution 4 (Line Diluent) are transferred to the cuvette of the RV and the blank intensity of the sample is measured. A second aliquot of the predilution mixture is transferred to the cuvette along with the Cortisol Fluorescein Tracer. Cortisol from the sample and the Cortisol Fluorescein Tracer compete for binding sites on the antibody molecule. The intensity of polarized fluorescent light is measured by the FPIA optical assembly.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

Study Details:

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

   • Reproducibility (Precision) Study: Three buffer-based panel members (1, 2, and 3) were assayed, in replicates of two, at two separate times per day, for 20 days. This means each panel member had 80 measurements (2 replicates * 2 times/day * 20 days).
   • Comparison Study:
     • 130 endogenous and cortisol-spiked serum and sodium heparin plasma specimens.
     • 150 endogenous and cortisol-spiked urine specimens.
   • Expected Values Study:
     • 50 serum specimens (AM and PM collections).
     • 49 urine specimens (24-hour collection).
   • Data Provenance: Not explicitly stated (e.g., country of origin). The studies appear to be prospective, as samples were "tested using the AxSYM Cortisol assay" and "collected from apparently healthy individuals" for the expected values study.

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

   • Not applicable as this is an in-vitro diagnostic (IVD) device measuring a biomarker, not interpreting imaging or clinical data that would require expert consensus. The "ground truth" for the comparison study is the result obtained from the predicate device (Beckman Access® Cortisol assay).

3. Adjudication method for the test set:

   • Not applicable. The "ground truth" for the comparison study is the result from the predicate device.

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 an IVD device, not an AI-based imaging or
     diagnostic device that involves human readers interpreting results.

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

   • Yes, the performance characteristics (reproducibility and comparison studies) reflect the standalone performance of the AxSYM® Cortisol assay without human-in-the-loop performance. The "human-in-the-loop" for this type of device would involve laboratory technicians performing the test, but the performance metrics provided are for the assay itself.

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

   • For the comparison study, the ground truth was the results obtained from the predicate device (Beckman Access® Cortisol assay).
   • For the reproducibility study, the "ground truth" is inherent to statistical precision analysis from repeated measurements.
   • For the expected values study, the "ground truth" was the cortisol levels measured by the device itself in apparently healthy individuals to establish reference ranges.

7. The sample size for the training set:

   • Not explicitly stated. For an IVD device, there isn't typically a "training set" in the same sense as machine learning algorithms. The development of the assay reagents and parameters would involve extensive R&D, but specific "training set" sizes are not usually disclosed in 510(k) summaries as they are not a discrete dataset used to train an algorithm.

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

   • Not applicable, as there's no explicitly defined "training set" with ground truth in the context of an IVD like there would be for an AI algorithm. The development process would rely on established biochemical principles and extensive internal testing/optimization.

Ask a specific question about this device

Immunological in vitro assay for the quantitative determination of cortisol in human serum, plasma and urine. The determination of cortisol is used for the recognition and treatment of functional disorders of the adrenal gland.

The electrochemiluminescence immunoassay "ECLIA" is intended for use on the Roche Elecsys® 1010 / 2010 and Modular Analytics E170 (Elecsys module) Immunoassay Analyzers.

The ELECSYS® Cortisol Assay a two step sandwich immunoassay with streptavidin microparticles and electrochemiluminescence detection. Results are determined using a calibration curve that is generated specifically on each instrument by a 2-point calibration and a master curve provided with the reagent bar code.

The information provided describes the Elecsys® Cortisol Assay and its substantial equivalence to a predicate device (Bayer Diagnostics ACS:180 Cortisol Assay). It focuses on comparing the new device's performance characteristics to the predicate device to demonstrate substantial equivalence, rather than detailing a specific study proving the device meets acceptance criteria with predefined thresholds. Therefore, some of the requested information cannot be fully extracted as it pertains to a typical clinical validation study with explicitly stated acceptance criteria and a detailed study design.

Here's the breakdown of what can be extracted and what is not available in the provided text:

1. Table of acceptance criteria and the reported device performance:

The document doesn't explicitly state "acceptance criteria" through specific numerical thresholds for each performance characteristic. Instead, it demonstrates performance by comparing the new device's data directly to that of the predicate device, implying that performance comparable to the legally marketed predicate device is the de facto acceptance criteria for substantial equivalence.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

This information is not provided in the given text. The document refers to "performance characteristics" but does not detail the specific study designs, sample sizes, or data provenance (e.g., retrospective/prospective, country of origin) for the tests conducted to derive these characteristics.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

This is a an in vitro diagnostic (IVD) device for quantitative determination of cortisol. The ground truth for such devices is typically established through analytical methods and reference materials, not through expert reading or consensus in the way a diagnostic imaging device would require. Therefore, this question is not applicable in the context of this device and information is not provided.

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

As this is an IVD device measuring an analyte, adjudication methods involving human experts (like 2+1 for radiology images) are not applicable. The ground truth is analytical.

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

This is an IVD assay, not an AI-powered diagnostic imaging device that assists human readers. Therefore, an MRMC comparative effectiveness study involving human readers and AI assistance is not applicable.

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

The entire device (ELECSYS® Cortisol Assay) operates as a standalone analytical system to quantitatively determine cortisol. The performance characteristics listed (precision, functional sensitivity, measuring range, etc.) represent the standalone performance of the algorithm and associated instrumentation.

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

For an in vitro diagnostic assay like the ELECSYS® Cortisol Assay, the "ground truth" for reported performance characteristics such as precision and sensitivity is established through analytical reference methods and certified reference materials. These involve meticulously prepared samples with known concentrations of cortisol, against which the device's measurements are compared. The document does not explicitly state the specific reference methods or materials used but implies their use through the reporting of these analytical performance metrics.

8. The sample size for the training set

This information is not provided. The document is a 510(k) submission for substantial equivalence, which typically focuses on performance data for the final product, not the developmental or training data used in its creation (especially for traditional IVD assays not based on machine learning in the way a modern AI device would be).

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

As per point 8, information about a "training set" and its ground truth establishment is not provided and likely not relevant in the context of this type of established IVD technology. The development of such assays involves rigorous analytical testing against reference standards, but this is distinct from the machine learning "training set" paradigm.

Ask a specific question about this device

For the quantitative determination of cortisol in serum using the Ciba Corning Automated Chemiluminescence Systems.

The Ciba Corning ACS Cortisol assay is a competitive chemiluminescent immunoassay. Cortisol in the patient sample competes with the acridinium ester (AE-labeled cortisol (Lite Reagent) for binding to polyclonal rabbit anti-cortisol antibody on the Solid Phase. The polyclonal rabbit anti-cortisol antibody is bound to monoclonal mouse anti-rabbit antibody, which is coupled to paramagnetic particles (solid phase). An inverse relationship exists between the amount of cortisol present in the patient sample and the amount of relative light units (RLUs) detected by the ACS:180® system.

The provided text describes a medical device, the Ciba Corning ACS Cortisol Immunoassay, and its performance data as part of a 510(k) summary. However, this document does not contain any information about acceptance criteria, clinical studies with human readers, or an AI device.

The document describes an immunoassay, which is a laboratory test that measures the presence or concentration of a substance (in this case, cortisol) using the reaction of an antibody to its antigen. This is a chemical/biological assay, not an AI or imaging device.

Therefore, I cannot provide the requested information, such as:

• A table of acceptance criteria and the reported device performance: This document reports sensitivity and accuracy against predicate devices, but it doesn't define "acceptance criteria" in the context of an AI-driven device with performance metrics like specificity, sensitivity, or AUC.
• Sample size used for the test set and the data provenance: While sample sizes for accuracy studies are given (100 extracted urine samples, 70 extracted urine samples), there's no mention of a "test set" in the context of AI. Data provenance is not specified beyond "patient sample" or "extracted urine samples."
• Number of experts used to establish the ground truth... and qualifications: Not applicable as this is not an AI/imaging device requiring expert interpretation for ground truth.
• Adjudication method: Not applicable.
• Multi reader multi case (MRMC) comparative effectiveness study: Not applicable, as there are no human readers or AI in this context.
• Standalone (i.e. algorithm only without human-in-the loop performance) was done: Not applicable.
• Type of ground truth used: For the accuracy section, the "ground truth" seems to be the results obtained by the predicate devices (Fluorescence Polarization Immunoassay and Radioimmunoassay).
• Sample size for the training set: Not applicable, as there is no AI model to train.
• How the ground truth for the training set was established: Not applicable.

In summary, the provided document describes a traditional in-vitro diagnostic immunoassay. It does not contain any information related to AI, machine learning, or clinical studies involving human readers or expert consensus for ground truth, which are the core components of your request.

Ask a specific question about this device