(255 days)
Trinity Biotech Uni-Gold™ Cryptosporidium is a single use rapid immunoassay for the qualitative detection of Cryptosporidium parvum (C. parvum) antigens in human stool specimens. This test is intended for use with patients with gastrointestinal symptoms as an aid in the diagnosis of suspected Cryptosporidium gastrointestinal infections. As with other Cryptosporidium tests, results should be considered in conjunction with the clinical evaluation and medical history. For In-Vitro Diagnostic use.
The Trinity Biotech Uni-Gold™ Cryptosporidium Test was designed as a single use, rapid, lateral flow immunoassay to detect the presence of Cryptosporidium parvum antigen in unpreserved (fresh & frozen), preserved, and media containing human stool specimens. The Trinity Biotech Uni-Gold™ Cryptosporidium test strip (5mm x 60mm) combines a nitrocellulose membrane with designated fiber pads (conjugate, sample and absorbant). The test strip is then placed into a plastic housing and is sealed constituting the Test Device.
Here's an analysis of the acceptance criteria and study details for the Uni-Gold™ Cryptosporidium device, based on the provided 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
The 510(k) summary does not explicitly state pre-defined "acceptance criteria" in terms of numerical thresholds for sensitivity and specificity. Instead, it presents the results of studies and aims to demonstrate substantial equivalence to a predicate device. For the purpose of this request, I will infer the implicit acceptance is achieving high sensitivity and specificity in comparison to gold standard methods and predicate devices.
| Metric / Study | Acceptance Criteria (Implicit) | Reported Device Performance (Uni-Gold™ Cryptosporidium) |
|---|---|---|
| Sensitivity (vs. DFA Microscopy) | High agreement with DFA microscopy | 100% (77/77) 95% CI 94 - 100% |
| Specificity (vs. DFA Microscopy) | High agreement with DFA microscopy | 100% (157/157) 95% Cl 97 - 100% |
| PPA (vs. Modified Acid-Fast Stain) | High positive percent agreement | 100% (26/26) |
| NPA (vs. Modified Acid-Fast Stain) | High negative percent agreement | 100% (21/21) |
| PPA (vs. Modified Kinyoun Stain) | High positive percent agreement | 92% (55/60) |
| NPA (vs. Modified Kinyoun Stain) | High negative percent agreement | 90% (198/221) |
| Specificity (Prospective Study vs. DFA Microscopy) | High specificity (as no positives were found) | 100% (378/378) 95% Cl 99 - 100% |
| Inter-run Precision/Reproducibility | 100% reproducibility across sites and days | 100% reproducibility observed for all sites, for all days |
| Cross-Reactivity | No cross-reactivity with common interfering organisms/substances | No cross-reactivity observed with listed organisms; No test interference observed by listed compounds |
2. Sample Size Used for the Test Set and Data Provenance
- Comparator Device Study (Side-by-side with Remel Xpect®): 299 retrospective samples. Data provenance: United States (multiple external sites). Sample matrices included unpreserved frozen, Cary Blair, SAF, and formalin.
- Sensitivity/Specificity Study (vs. DFA Microscopy): 234 retrospective samples (77 positive, 157 negative). Data provenance: United States (2 external laboratories). Sample matrices included formalin, SAF, unpreserved frozen, Cary Blair, and C&S.
- Additional Retrospective Studies (vs. Non-fluorescent Microscopy):
- Site 2: 47 retrospective samples (26 positive, 21 negative) vs. Modified Acid-Fast Stain.
- Site 3: 281 retrospective SAF samples (60 positive, 221 negative) vs. Modified Kinyoun Stain.
- Prospective Study (vs. DFA Microscopy): 378 samples (all negative). Data provenance: one external laboratory in the US. Sample matrices included unpreserved fresh, unpreserved frozen, formalin, SAF, C&S, and Cary Blair.
- Overall Combined Test Set: 940 (prospective and retrospective) samples. Data provenance: Collected from Hospitals throughout the US and Canada.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The document does not specify the number of individual experts or their exact qualifications (e.g., number of years of experience, specific medical specialty) used to establish the ground truth for the test sets.
Instead, it refers to "DFA microscopy," "Modified Kinyoun Stain light microscopy," and "Modified Acid-Fast Stain" as the reference methods (ground truth). It is implied that these microscopy methods are performed and interpreted by trained laboratory professionals, but specific expert details are not provided.
4. Adjudication Method for the Test Set
The document does not describe an explicit adjudication method (e.g., 2+1 reader consensus) for resolving discrepancies in the reference methods (DFA microscopy, Modified Kinyoun Stain, Modified Acid-Fast Stain).
In cases of discrepancy between the Uni-Gold™ device and the comparator/reference method within the studies, some retesting or re-evaluation against a "higher" ground truth was performed:
- Site 3 (Comparator Device Study): For the 51 samples positive by Uni-Gold™ but negative by comparator, 30 were positive by Modified Kinyoun Stain light microscopy and 3 by DFA microscopy (agreeing with Uni-Gold™). The one sample negative by Uni-Gold™ and positive by comparator was negative by Modified Kinyoun Stain microscopy (agreeing with Uni-Gold™). This indicates that microscopy was used as an adjudicator for samples where the test device and the comparator device disagreed.
- Site 3 (Non-fluorescent Microscopy Study): For the 23 negative samples (by Modified Kinyoun Stain) that tested positive on Uni-Gold™, three subsequently tested positive by DFA microscopy (agreeing with Uni-Gold™). This implies DFA microscopy acted as a higher-level adjudicator for certain discrepancies.
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 question is not applicable to the Uni-Gold™ Cryptosporidium device. The device is a rapid in vitro diagnostic immunoassay, not an AI-assisted diagnostic tool that involves human readers interpreting images. Therefore, an MRMC comparative effectiveness study involving human readers and AI assistance was not performed and is not relevant to this type of device.
6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done
This question is also not applicable. The Uni-Gold™ Cryptosporidium device is a standalone lateral flow immunoassay. Its performance is the "algorithm only" performance (the chemical reaction and visual readout) without human intervention in the detection process itself, beyond the initial sample preparation and result interpretation. There is no separate "algorithm without human-in-the-loop" performance as it is the test.
7. The Type of Ground Truth Used
The primary types of ground truth used for the test sets were:
- DFA Microscopy: Direct Fluorescent Antibody microscopy, considered a gold standard for Cryptosporidium detection.
- Modified Kinyoun Stain Light Microscopy: A non-fluorescent microscopy staining method.
- Modified Acid-Fast Stain: Another non-fluorescent microscopy staining method.
- Comparator Device Results: In some comparative studies, the results of the legally marketed predicate device (Remel Xpect® Cryptosporidium) were used for comparison. However, when discrepancies arose in these studies, microscopy methods (DFA or Modified Kinyoun) were often used to adjudicate.
- Clinical Evaluation and Medical History: The intended use statement notes that "results should be considered in conjunction with the clinical evaluation and medical history," implying that clinical context is part of the broader diagnostic process, but the technical ground truth for device performance was microscopy.
8. The Sample Size for the Training Set
The document does not explicitly mention a "training set" in the context of device development or any machine learning algorithm. This device is a biochemical immunoassay, not an AI or machine learning model that requires a discrete training set. The development of the assay's reagents and parameters would have involved internal validation and optimization, but not in the sense of a machine learning training set.
9. How the Ground Truth for the Training Set Was Established
Since there is no "training set" for an AI/ML algorithm in this context, this question is not applicable. The ground truth for developing and optimizing the immunoassay would have been established through standard laboratory practices using known positive and negative controls, and possibly characterized clinical samples, during the assay development process, prior to the formal validation studies presented here.
§ 866.3220
Entamoeba histolytica serological reagents.(a)
Identification. Entamoeba histolytica serological reagents are devices that consist of antigens and antisera used in serological tests to identify antibodies toEntamoeba histolytica in serum. Additionally, some of these reagents consist of antisera conjugated with a fluorescent dye (immunofluorescent reagents) used to identifyEntamoeba histolytica directly from clinical specimens. The identification aids in the diagnosis of amebiasis caused by the microscopic protozoan parasiteEntamoeba histolytica and provides epidemiological information on diseases caused by this parasite. The parasite may invade the skin, liver, intestines, lungs, and diaphragm, causing disease conditions such as indolent ulcers, an amebic hepatitis, amebic dysentery, and pulmonary lesions.(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 § 866.9.