Search Results

A1C EZ Glycohemoglobin Analysis System is an in vitro diagnostic test used to quantitatively measure the percent glycohemoglobin Alc or glycohemoglobin A1c fraction mmol/mol in venous whole blood samples. This system is intended for multiple patient use to monitor long term glycemic control in individuals previously diagnosed with diabetes. This test is not to be used for screening or diagnosis of diabetes or for neonatal use. The A1C EZ Glycohemoglobin Analysis System is intended for professional use in clinical laboratories only.

This product is used to quantitatively measure the percent glycohemoglobin A1c or glycohemoglobin A1c fraction mmol/mol in human whole blood sample.

Here is an organized overview of the acceptance criteria and study details for the A1C EZ Glycohemoglobin Analysis System:

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

The document does not explicitly present a table of acceptance criteria. However, it details various performance characteristics and compares them to a predicate device. I've extracted the key performance metrics from the non-clinical testing section and the "Substantial Equivalence Comparison" table, treating them as implicitly accepted performance 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)

• Test Sets for Precision:
  • Internal Precision: "Sample from normal patient (Level 1, HbA1c% = 5.8%) and diabetic patient (Level 2, HbA1c% = 12.2%)" were used. The study involved running samples over 20 consecutive days. The exact number of individual patient samples (N) is not explicitly stated, but rather referred to as "Level 1" and "Level 2" control samples.
  • External Precision: "3 levels of K2-EDTA anticoagulated venous blood samples (level 1- 5.1%, level 2- 7.3%, level 3- 11.7%)" were used. For each sample level, 360 tests were performed across all sites, operators, meters, and test strip lots (N=360 for each level).
• Test Sets for Accuracy: The "internal comparison study" does not specify the number of individual patient samples. It mentions comparison with a reference device (Tosoh HLC-723 G8) and the predicate device (A1CNow+), implying a set of samples with known values.
• Test Sets for Linearity: Not specified.
• Test Sets for Analytical Specificity: Not specified, but involved testing various interferent substances at specified concentrations.
• Data Provenance: The document does not explicitly state the country of origin for the patient samples or if the data was retrospective or prospective. Given the applicant is from China, it is plausible the studies were conducted in China.

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 device is an in vitro diagnostic (IVD) quantitative measurement system, not an imaging device requiring expert interpretation. Therefore, the concept of "experts" to establish ground truth in the traditional sense (e.g., radiologists, pathologists) is not applicable here.

Instead, the ground truth for performance evaluation (accuracy) was established by comparison to a recognized reference method:

• Reference Method: Tosoh HLC-723 G8, which is an HPLC device certified by the National Glycohemoglobin Standardization Program (NGSP) and International Federation of Clinical Chemistry and Laboratory Medicine (IFCC). This method is considered the "gold standard" for HbA1c measurement.

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

Not applicable. As an IVD device measuring a quantitative biomarker, adjudication by multiple human readers is not a part of the evaluation process. The accuracy is assessed by comparing the device's numerical output to a highly accurate reference method.

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 IVD device for quantitative measurement, not an AI-assisted diagnostic tool that involves human readers interpreting cases.

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

Yes, the studies described are for the standalone performance of the A1C EZ Glycohemoglobin Analysis System. It's an automated system that provides a quantitative measurement, inherently operating without human-in-the-loop performance influencing the measurement itself. Users operate the device, but their interpretation skills are not part of the device's core performance evaluation here; rather, the device's ability to accurately measure HbA1c is what's being assessed.

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

The ground truth for the device's accuracy was established by comparison to a reference method – the Tosoh HLC-723 G8 HPLC device. This method is internationally recognized and certified (NGSP and IFCC), serving as the de facto "truth" for HbA1c concentration.

8. The sample size for the training set

This document does not describe a "training set" in the context of machine learning (AI). This is a traditional IVD device, and the listed studies are for its analytical performance validation rather than AI model training. Therefore, a training set as understood in AI/ML is not applicable.

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

As there is no "training set" in the AI/ML sense for this device, this question is not applicable. The device's underlying chemistry and photometric measurement principles are based on established scientific methods for HbA1c detection, not statistical learning from a labeled training dataset.

Ask a specific question about this device