FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The Anti-Borrelia burgdorferi US EUROLINE-WB (IgM) kit is a Western blot assay intended for the qualitative determination of IgM class antibodies against Borrelia burgdorferi in human serum and plasma (K+-EDTA, Li+-heparin, Na+-citrate) samples that have been found positive or equivocal/borderline using an enzyme immunoassay (EIA) or immunofluorescence assay (IFA) test procedure for Borrelia burgdorferi antibodies. Results can be read manually or automated utilizing EUROLineScan. This test is used as an aid in the diagnosis of infections with Borrelia burgdorferi and the associated diseases, in conjunction with other laboratory and clinical findings.

Device Description

AI/ML Overview

This is an FDA 510(k) clearance letter for the "Anti-Borrelia burgdorferi US EUROLINE-WB (IgM)" device, which is an in vitro diagnostic (IVD) test. IVDs are not typically "AI" devices as understood in the context of machine learning algorithms for image analysis or other complex data interpretation. Therefore, many of the requested fields related to AI-specific studies (e.g., sample size for test set split, MRMC study, training set details) are not applicable to this type of device.

However, I can extract information related to its analytical performance, which is a form of "acceptance criteria" and "device performance" for IVDs.

Device Name: Anti-Borrelia burgdorferi US EUROLINE-WB (IgM)

Indications for Use: The kit is a Western blot assay for qualitative determination of IgM class antibodies against Borrelia burgdorferi in human serum and plasma samples. It is intended for samples that have tested positive or equivocal/borderline by EIA or IFA for Borrelia burgdorferi antibodies. It is used as an aid in the diagnosis of Borrelia burgdorferi infections in conjunction with other laboratory and clinical findings.

Given the nature of an IVD, the "acceptance criteria" generally refer to established performance benchmarks (e.g., sensitivity, specificity, agreement with a reference method) that demonstrate the device is safe and effective for its intended use. The "study" would be the clinical performance evaluation or method comparison study.

Since the provided document is a 510(k) clearance letter, it typically summarizes the decision but does not include the full study report. However, based on the context of IVD submissions, I can infer the type of data that would have been presented for acceptance.

1. Table of Acceptance Criteria and Reported Device Performance

For an IVD like this, acceptance criteria would typically involve demonstrating substantial equivalence to a predicate device or meeting certain performance specifications like sensitivity and specificity against a "gold standard" or well-characterized samples. The exact numerical acceptance criteria are not in this document, but I can describe the type of performance metrics expected.

Metric (Type of Acceptance Criteria)	Reported Device Performance (Likely presented in full submission)
PPA (Positive Percent Agreement)	High agreement with a comparator method or known positive samples, indicating good sensitivity. Specific value not in document.
NPA (Negative Percent Agreement)	High agreement with a comparator method or known negative samples, indicating good specificity. Specific value not in document.
Overall Agreement	High concordance with a predicate device or reference method. Specific value not in document.
Precision/Reproducibility	Consistent results across different runs, operators, and sites. (Not explicitly in clearance letter, but expected for IVD).
Cross-reactivity	Lack of false positives with antibodies to other pathogens. (Not explicitly in clearance letter, but expected for IVD).
Interference	Lack of impact from endogenous or exogenous substances. (Not explicitly in clearance letter, but expected for IVD).

Note: The specific numerical values for PPA, NPA, and Overall Agreement would be found in the detailed study report submitted with the 510(k), which is not part of this clearance letter. The clearance letter only states that the device was found "substantially equivalent" to predicate devices based on the submitted performance data.

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

Sample Size for Test Set: Not explicitly stated in the clearance letter. IVD studies typically involve hundreds to over a thousand samples for clinical performance.
Data Provenance: Not explicitly stated. For a US FDA submission, data often includes samples from diverse geographic regions within the US, and sometimes international samples if relevant to the intended use population. It's highly likely to be retrospective (using archived, characterized samples) but could include prospective samples as well.

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

Number of Experts/Qualifications: This concept of "number of experts" for ground truth is generally not applicable in the same way for an IVD diagnostic kit that measures antibodies. The "ground truth" for such a device is established by:
- Clinical Diagnosis: Patient's diagnosis by a physician based on clinical symptoms, history, and other laboratory tests.
- Reference Tests: Results from established, FDA-cleared/approved predicate devices or a panel of standard reference tests (e.g., PCR, culture, or a combination of multiple serological tests).
- Well-Characterized Sample Panels: Samples from patient cohorts definitively diagnosed as positive or negative for the condition being tested, confirmed by multiple methods.

4. Adjudication Method for the Test Set

Adjudication Method: Not applicable in the traditional sense for an IVD like this. The "ground truth" is determined by the methods described in point 3, not by expert consensus on image interpretation. If there were discrepancies between the new device and the reference method, these would be investigated, but not typically "adjudicated" by experts in the sense of image review.

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 (IVD) kit for antibody detection, not an AI-powered diagnostic imaging device or an AI human-in-the-loop system. The device itself performs the detection (manually or with an automated reader, EUROLineScan), and the result is interpreted. There are no human "readers" whose performance is being improved by "AI assistance" in the context of this device.

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

Yes, this is essentially a standalone device. The device itself (the Western blot assay) generates a result (presence/absence of specific IgM bands), which can be read manually or automatically by EUROLineScan. This is the "algorithm only" performance for the assay itself. The performance provided in the 510(k) submission would represent the standalone performance of the test. While a human might manually interpret the bands, the core performance metrics are about the assay's ability to correctly identify antibodies, not a human's ability to interpret an image aided by AI.

7. The Type of Ground Truth Used

Type of Ground Truth: Most likely a combination of:
- Clinical Diagnosis: Based on physician assessment, patient history, and clinical presentation of Lyme disease.
- Reference Laboratory Testing: Results from established, often predicate, serological assays (e.g., EIA/IFA followed by another Western Blot) and potentially other confirmatory tests (though less common for Borrelia serology).
- Well-characterized Patient Samples/Panels: Samples obtained from individuals with confirmed Lyme disease and healthy controls, often with long-standing clinical diagnoses and/or multiple confirmatory tests.

8. The Sample Size for the Training Set

Not Applicable in the context of machine learning algorithms. For an IVD kit like this, there isn't a "training set" in the sense of data used to train a machine learning model. The device performance characteristics are established through analytical validation and clinical performance studies on distinct panels of samples.

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

Not Applicable. See point 8. The "ground truth" for validating the device's performance (the "test set" in ML terms) would be established as described in point 7. For the development of the assay itself (e.g., selecting antigens, optimizing reagents), manufacturers use internal development panels, but these are not referred to as "training sets" with "ground truth" in the machine learning sense for regulatory submissions of IVDs.

Summary

Regulation Number and Section

§ 866.3830

Treponema pallidum treponemal test reagents.(a)
Identification. Treponema pallidum treponemal test reagents are devices that consist of the antigens, antisera and all control reagents (standardized reagents with which test results are compared) which are derived from treponemal sources and that are used in the fluorescent treponemal antibody absorption test (FTA-ABS), theTreponema pallidum immobilization test (T.P.I.), and other treponemal tests used to identify antibodies toTreponema pallidum directly from infecting treponemal organisms in serum. The identification aids in the diagnosis of syphilis caused by bacteria belonging to the genusTreponema and provides epidemiological information on syphilis.(b)
Classification. Class II (performance standards).