(60 days)
The ACS:180 and ADVIA Centaur Valproic Acid Immunoassays are competitive, chemiluminescence immunoassay for the quantitative determination of valproic acid in human serum and plasma for use on the automated analyzers marketed by Bayer Corporation. Valproic acid (2-propylpentanoic acid) is an anticonvulsant that is used alone or in combination with other anticonvulsant drugs to control seizures. Monitoring of valproic acid ensures that there is adequate drug in the blood stream without being at toxic levels. The ACS:180 and ADVIA Centaur Valproic Acid Immunoassays are used as an aid to monitor patients' valproic acid level.
The ACS and ADVIA Centaur Valproic acid assay is a competitive chemiluminescence immunoassay intended for the quantitative determination of valproic acid in human serum and plasma. Valproic acid in the patient sample, calibrators, standards and controls competes with acridinium ester-labeled valproic acid in the Lite Reagent for a limited amount of monoclonal mouse anti-valproic acid antibody, which is covalently coupled to paramagnetic particles in the Solid Phase. Following incubation, non-reacted acridinium ester-labeled valproic acid and non-reacted valproic acid from the sample is washed from the reaction mixture. The chemiluminescence of the bound, labeled valproic acid is measured in a luminometer. The measured chemiluminescence is inversely proportional to the quantity of valproic acid in the sample.
The document describes a 510(k) premarket notification for the ADVIA Centaur and ACS:180 Valproic Acid Immunoassays. This is not a study that uses AI. The acceptance criteria and the study presented are for establishing substantial equivalence to a predicate device, the TDx Valproic acid assay, based on method comparison using correlation studies.
Here's an analysis of the provided information within the context of your request, noting where the information is not applicable due to the nature of the device (an immunoassay, not an AI device):
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for substantial equivalence are inferred from the correlation study results, which show a strong linear relationship and high correlation coefficient between the new devices and the predicate. While explicit numerical acceptance criteria (e.g., "slope between 0.9 and 1.1") are not directly stated as "acceptance criteria," the reported performance demonstrates that these criteria were met for substantial equivalence.
| Acceptance Criteria (Inferred for Substantial Equivalence via Method Comparison) | Reported Device Performance (ADVIA Centaur Valproic acid vs. TDx valproic acid) | Reported Device Performance (ACS:180 Valproic acid vs. TDx valproic acid) | Reported Device Performance (ADVIA Centaur Valproic acid vs. ACS:180 Valproic acid) |
|---|---|---|---|
| Slope close to 1.0 | 0.96 | 0.98 | 0.97 |
| Intercept close to 0.0 | 4.03 | 1.37 | 3.37 |
| Correlation Coefficient (r) close to 1.0 (indicating strong linear relationship) | 0.99 | 0.99 | 0.99 |
2. Sample Sizes Used for the Test Set and Data Provenance
The document refers to the data as "correlation studies" and does not explicitly differentiate between "test set" and "training set" in the context of machine learning. Thus, the sample sizes provided are for the datasets used in these correlation studies.
- Sample sizes for correlation studies (acting as test sets):
- ADVIA Centaur Valproic acid vs. TDx valproic acid: N = 250
- ACS:180 Valproic acid vs. TDx valproic acid: N = 250
- ADVIA Centaur Valproic acid vs. ACS:180 Valproic acid: N = 253
- Data Provenance: Not specified. It's common for such studies to use clinical samples, but the country of origin or whether they were retrospective/prospective is not mentioned.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
This information is not applicable as the ground truth for an immunoassay is typically established by the reference method (the predicate device, TDx Valproic acid assay) or by laboratory standards and calibrators, not by expert human graders or consensus. The "ground truth" here is the measurement obtained from the established method.
4. Adjudication Method for the Test Set
This information is not applicable. Adjudication methods (like 2+1, 3+1) are used for resolving discrepancies among human experts (e.g., radiologists, pathologists) who are establishing ground truth for subjective interpretations, especially in AI studies. For an immunoassay, the "ground truth" is a quantitative measurement, and disputes are resolved through quality control, calibration, and repeat testing, not expert adjudication in the traditional sense.
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 information is not applicable. The device is a quantitative immunoassay, not an AI diagnostic tool that assists human readers. Therefore, an MRMC study and analysis of human reader improvement with AI assistance were not performed.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
This information is not applicable in the context of an AI algorithm. The performance presented is inherently "standalone" in the sense that it's the performance of the immunoassay itself in comparison to a predicate, not an AI algorithm. There is no human-in-the-loop component for reading/interpreting the immunoassay results that would be "assisted" by AI.
7. The Type of Ground Truth Used
The "ground truth" for the comparison studies was established by the predicate device, the TDx Valproic acid assay. This means the new devices (ADVIA Centaur and ACS:180) were compared against an already legally marketed and accepted method for measuring valproic acid.
8. The Sample Size for the Training Set
The document does not mention a "training set" in the context of algorithm development, as this is an immunoassay, not an AI-driven device. The term "training set" is typically used for machine learning. The studies described are method comparison studies.
9. How the Ground Truth for the Training Set Was Established
This information is not applicable as there is no "training set" for an AI algorithm described in the document.
§ 862.3200 Clinical toxicology calibrator.
(a)
Identification. A clinical toxicology calibrator is a device intended for medical purposes for use in a test system to establish points of reference that are used in the determination of values in the measurement of substances in human specimens. A clinical toxicology calibrator can be a mixture of drugs or a specific material for a particular drug (e.g., ethanol, lidocaine, etc.). (See also § 862.2 in this part.)(b)
Classification. Class II (special controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.