Search Results

The Access TSH (3rd IS) assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of human thyroid-stimulating hormone (thyrotropin, TSH, hTSH) levels in human serum and plasma using the Access Immunoassay Systems. This assay is capable of providing 3rd generation TSH results.

The Access TSH (3rd IS) Calibrators are intended to calibrate the Access TSH (3rd IS) assay for the quantitative determination of human thyroid-stimulating hormone (thyrotropin, TSH, hTSH) levels in human serum and plasma using the Access Immunoassay Systems.

The Access TSH (3rd IS) Assay (standardized to WHO 3rd International Standard, 81/565), Access TSH (3rd IS) Calibrators, and the Access Immunoassay analyzers comprise the Access Immunoassay System for the quantitative determination of thyroid-stimulating hormone (thyrotropin, TSH, hTSH) in human serum and plasma.

The provided text describes a 510(k) premarket notification for a medical device: "Access TSH (3rd IS) Assay and Access TSH (3rd IS) Calibrators on the Access Immunoassay Systems." This document primarily focuses on demonstrating substantial equivalence to a predicate device rather than presenting a study design and results against specific acceptance criteria for a novel AI/ML-driven diagnostic device.

Therefore, many of the typical acceptance criteria and study design elements requested in your prompt (e.g., sample sizes for training/test sets in AI/ML, number of experts for ground truth, MRMC studies, standalone algorithm performance) are not directly applicable or discussed in this type of submission. This document describes a traditional in-vitro diagnostic (IVD) assay, not an AI/ML device.

However, I can extract the closest equivalent information based on the provided text, interpreting "acceptance criteria" as performance specifications and "study" as the validation studies performed for this IVD device.

Acceptance Criteria and Reported Device Performance (Table)

While formal "acceptance criteria" are not presented as such for an AI/ML model, the document details various performance characteristics and their achieved results. The "Predicate" column can be seen as a de facto benchmark or comparison point for substantial equivalence.

Study Details for the Access TSH (3rd IS) Assay:

1. Sample Size used for the Test Set and Data Provenance:

   • Method Comparison: 155 serum samples.
   • Imprecision: Not explicitly stated, but implies multiple replicates of various serum samples across TSH concentrations.
   • High-dose Hook Effect: One serum sample tested up to 1,000 µIU/mL.
   • Linearity & Dilution Recovery: Serum samples (quantity not specified for the overall study design of these tests, but likely multiple samples at various levels).
   • LoB, LoD, LoQ: Serum samples (quantity not specified).
   • Analytical Specificity: Serum samples (for interference), and two individual patient serum samples for cross-reactivity.
   • Expected Reference Intervals:
     • General Population (non-pregnant females and males): 367 subjects.
     • Pregnant females - 1st Trimester: 318 subjects.
     • Pregnant females - 2nd Trimester: 362 subjects.
     • Pregnant females - 3rd Trimester: 335 subjects.
   • Matrix Comparison: 79 matched sets of serum (gel and no gel) and plasma (lithium-heparin) samples.
   • Data Provenance: The document does not specify the country of origin of the data or whether it was retrospective or prospective. Given the nature of IVD studies, it's typically prospective collection for performance validation.

2. Number of Experts used to establish the Ground Truth for the Test Set and Qualifications of those Experts:

   • Not applicable in the context of an IVD assay measuring an analyte concentration. The "ground truth" for TSH levels is established by the reference measurement procedure or the predicate device's measured values, rather than expert interpretation of images or other subjective data. For reference interval studies, the "ground truth" would be the clinical status (euthyroid, pregnant trimester) determined by a clinician, not an "expert."

3. Adjudication Method for the Test Set:

   • Not applicable as this is a quantitative chemical assay, not an interpretative diagnostic device requiring human adjudication of results.

4. If a multi-reader multi-case (MRMC) comparative effectiveness study was done:

   • No. This type of study is relevant for AI/ML devices where human readers' performance is augmented by AI. This document pertains to an automated IVD assay.

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

   • Yes, implicitly. The performance metrics (imprecision, linearity, LoD, etc.) are all measures of the device's standalone analytical performance without human intervention in the result generation. The "method comparison" compares the new device's output to the predicate device's output.

6. The type of ground truth used:

   • Quantitative Reference Measurement: For analytical performance characteristics (e.g., limit of detection, linearity, imprecision), the ground truth is based on gravimetric/volumetric preparation of controls and calibrators, or comparison to a well-characterized predicate device (which itself has its own reference measurement traceabilities).
   • Clinical Status: For the establishment of "Expected Reference Intervals," the ground truth is the clinical classification of the subjects (e.g., "apparently healthy, normal euthyroid," "pregnant females - 1st Trimester") presumably determined by medical assessment.
   • WHO International Reference Preparations: The calibrators and assay are standardized to specific WHO International Reference Preparations (e.g., WHO 3rd International Reference Preparation Thyroid Stimulating Hormone, Human, NIBSC Coded 81/565). This is the ultimate "ground truth" for the quantitative value of TSH.

7. The sample size for the training set:

   • Not applicable. This is an IVD assay, not an AI/ML device that requires a training set in the typical sense. The "training" for such a device occurs during its development and optimization, not as a distinct "training set" like in machine learning. The calibration process uses specific calibrator levels to generate a calibration curve.

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

   • Not applicable for the same reason as above. The "ground truth" for calibration is derived from the assigned values of the WHO International Reference Preparations through the calibrators.

Ask a specific question about this device