(347 days)
The S-Test Albumin Reagent is intended for the quantitative determination of albumin concentration in serum or heparin plasma using the S40 Clinical Analyzer. Albumin measurements are used in the diagnosis and treatment of numerous diseases involving primarily the liver or kidneys. This test is intended for use in clinical laboratories or physician office laboratories. For in vitro diagnostic use only.
The S-Test albumin (ALB) reagent cartridge used with the S40 Clinical Analyzer is intended for quantitative in vitro diagnostic determination of ALB in serum or heparin plasma based on a photometric test measuring the formation of a bluish-green complex from ALB and bromcresol green.
The S-Test ALB device, used with the S40 Clinical Analyzer, is an in vitro diagnostic intended for the quantitative determination of albumin concentration in serum or heparin plasma. The device's performance was evaluated through precision, accuracy, and sensitivity studies.
1. Acceptance Criteria and Reported Device Performance
| Parameter | Acceptance Criteria | Reported Device Performance |
|---|---|---|
| Precision | Not explicitly stated as acceptance criteria, but based on the reported data, the device should exhibit consistent reproducibility. | In-house (22 days, three ALB levels): Within-run CV ranged from 1.8 to 2.0%; total CV ranged from 4.7 to 5.6%. |
| POL sites (three sites, five days): Within-run CVs ranged from 0.6 to 2.5%; total CVs ranged from 0.8 to 2.8%. | ||
| Accuracy | Not explicitly stated as acceptance criteria, but based on the reported data, the device should demonstrate a strong correlation with a comparative method and acceptable regression analysis parameters. | Correlation study (92 samples, ALB 0.7 to 6.8 g/dL): Correlation coefficient (r) = 0.961; standard error estimate (SEE) = 0.24; confidence interval slope = 0.873 to 0.982; confidence interval intercept = 0.05 to 0.51. |
| Patient correlation studies (four POL sites): Correlation coefficients ranged from 0.987 to 0.994; standard error estimates ranged from 0.12 to 0.19; confidence interval slopes ranged from 0.942 to 1.040; confidence interval intercept ranged from -0.27 to 0.09. | ||
| Sensitivity | Not explicitly stated as acceptance criteria, but refers to the lowest detectable concentration. | The detection limit was 0.4 g/dL. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size for Accuracy (Correlation Study): 92 samples.
- Sample Size for Accuracy (Patient Correlation Studies): Not explicitly stated, but conducted at four separate Physician Office Laboratory (POL) sites, implying multiple patient samples from each site.
- Sample Size for Precision: Three ALB levels for 22 days (in-house) and three separate POL sites over five days. The number of individual samples per level or site is not specified.
- Data Provenance: The studies were conducted at "three separate Physician Office Laboratory (POL) sites" and "four separate POL sites," along with "in-house" testing. This suggests a mix of internal testing and external validation in real-world clinical settings (Physician Office Laboratories). The country of origin is not explicitly stated, but given the submission is to the FDA, it is likely the studies were conducted in the United States or under comparable regulatory standards. The data is prospective as it describes the performance of the device in studies.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications
The ground truth for this device is based on quantitative measurements of albumin concentration, not expert interpretation of qualitative data (e.g., images). Therefore, clinical experts (like radiologists) for establishing "ground truth" in the traditional sense are not applicable here. The "ground truth" is established by a "comparative method" or a reference method for albumin measurement.
4. Adjudication Method for the Test Set
Adjudication methods (like 2+1, 3+1) are typically used for qualitative assessments where there might be disagreement among human readers. Since the performance studies involve quantitative measurements of albumin concentration against a comparative method, adjudication methods are not relevant or described. The comparison is between the S-Test ALB measurements and the measurements from the comparative method.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, an MRMC comparative effectiveness study was not done. This type of study is relevant for evaluating the impact of AI on human reader performance for tasks involving interpretation (e.g., radiology). The S-Test ALB is a clinical analyzer for quantitative biochemical measurements, not an image interpretation or diagnostic aid for human readers.
6. Standalone Performance Study
Yes, a standalone study was done. The performance data presented (precision, accuracy, sensitivity) directly reflects the performance of the S-Test ALB reagent cartridge used with the S40 Clinical Analyzer as a standalone system. The results are based on the device's measurements compared to a "comparative method," which serves as the reference, rather than comparing it to human interpretation.
7. Type of Ground Truth Used
The ground truth used is quantitative measurements from a comparative method. For accuracy studies, the S40 Clinical Analyzer using S-Test ALB (y) was compared against a "comparative method" (x) for albumin determination. This implies that results from an established, independently validated method were used as the reference standard.
8. Sample Size for the Training Set
The document does not explicitly mention a "training set" or "training data" for the device. Clinical analyzers like the S40 Clinical Analyzer with S-Test ALB typically rely on established chemical reactions and calibration procedures rather than machine learning models that are "trained" on a dataset in the same way an AI algorithm would be. The "training" in this context would refer to instrument calibration and reagent validation.
9. How the Ground Truth for the Training Set Was Established
As noted above, a "training set" in the context of an AI algorithm is not applicable here. The "training" for a clinical analyzer system would involve calibration using known standards according to the manufacturer's protocols and quality control procedures. These standards would have their values established through highly accurate and traceable reference methods. The document does not describe the specific calibration or quality control protocols used, but these are standard practices for in vitro diagnostic devices.
§ 862.1035 Albumin test system.
(a)
Identification. An albumin test system is a device intended to measure the albumin concentration in serum and plasma. Albumin measurements are used in the diagnosis and treatment of numerous diseases involving primarily the liver or kidneys.(b)
Classification. Class II.