Search Results

The RIDASCREEN® Norovirus 3td Generation test is a qualitative enzyme immunoassay (EIA) intended for the detection of selected genogroup I (GI.1, GL.2, GI.3, GI.4, GL.7) and genogroup II (GII.1, GII.2, GII.3, GII.4, GII.5, GII.6, GII.7, GII.8, GII.10, GII.12, GII.13, GII.14, GII.17) norovirus strains in human feces as an aid in investigating the cause of acute gastroenteritis outbreaks. The likelihood of detecting a norovirus outbreak by use of the RIDASCREEN® Norovirus 3td Generation test improves as the number of patients tested during an outbreak increases, as well as when quality of the specimens increases. Preliminary identification of norovirus as the cause of an acute gastroenteritis outbreak by RIDASCREEN® Norovirus 3td Generation testing should be confirmed by reference methods as appropriate, particularly if only a limited number of positive samples are associated with a suspected outbreak. Additional testing of negative samples by other methods should be performed if norovirus is strongly suspected as the cause of an acute gastroenteritis outbreak.

RIDASCREEN® Norovirus 3rd Generation test is a solid phase sandwich-type EIA for the detection of genogroups GI and GII noroviruses in stool samples. Microwell strips are coated with a mixture of GI and GII norovirus specific monoclonal antibodies. An aliquot of fecal suspension is added to the microwell together with biotinylated monoclonal norovirus antibodies. After washing, streptavidin peroxidase conjugate is added. Norovirus antigens that are present in the stool sample are captured in a sandwich complex of the immobilized antibodies, the norovirus antigens and the monoclonal antibodies conjugated with the biotin-streptavidin-peroxidase complex. Unbound streptavidin peroxidase conjugate is removed by washing and a chromogenic colorless substrate solution (hydrogen peroxide/TMB) is added. The substrate is hydrolyzed by any bound peroxidase, changing the chromogen to a blue color. Stopping the reaction with acid converts the blue to a yellow color indicating the presence of norovirus antigens.

Test results are read photometrically; intensity significantly above background levels is indicative of the presence of Norovirus antigen in the specimen or control. The RIDASCREEN® Norovirus 3rd Generation test does not identify specific norovirus strains.

Here's an analysis of the provided text, extracting the requested information about acceptance criteria and the device study:

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

The document does not explicitly state formal "acceptance criteria" against which the device's performance was measured. Instead, it reports the device's performance characteristics. If we were to infer acceptance criteria from the reported performance in the Overall Clinical Trial, a reasonable interpretation would be the achieved positive agreement and negative agreement.

Inferred Acceptance Criteria (based on reported performance):

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Sample Size (Test Set): 609 samples (after excluding 69 invalid samples and 2 equivocal samples from an initial collection of 680).
• Data Provenance:
  • Country of Origin: Three sites in the United States (Ohio, California, Cincinnati) and one site in Canada (Hamilton, Ontario).
  • Retrospective or Prospective: Prospective (samples were collected from patients presenting with acute gastroenteritis symptoms).

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)

The ground truth for the test set was established using a "Clinical Diagnostic Truth" algorithm based on a composite reference endpoint, not directly by experts. The reference methods included:

• Quantitative real-time RT-PCR (qRT-PCR)
• Conventional RT-PCR with bi-directional sequencing (Region D and Region B)
• Electron Microscopy (EM)

While these methods would be performed by trained laboratory personnel, the document does not specify the number or qualifications of experts (e.g., virologists, molecular biologists) who established the final "Clinical Diagnostic Truth." However, the reference testing was conducted at established public health and disease control laboratories: the California Dept. of Public Health (CPDH) and the National Calicivirus Laboratory, Centers for Disease Control and Prevention (CDC).

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

The ground truth was established by a rule-based algorithm, not an adjudication process involving multiple human readers in the traditional sense. The algorithm for "Clinical Diagnostic Truth" considered a sample positive if it was positive by:

• Region D RT-PCR OR
• Region B RT-PCR OR
• Electron Microscopy (EM)

If all three methods were negative, the sample was considered negative. There was an initial "Original Composite 'Clinical Diagnostic Truth' Algorithm" that included qRT-PCR, but this was revised due to concerns, and the final algorithm focused on Region D RT-PCR, Region B RT-PCR, and EM.

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

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. This device is a diagnostic immunoassay for Norovirus detection, not an AI-assisted diagnostic tool designed to be used by human readers (like a radiologist reading an AI-analyzed image). Therefore, the concept of human readers improving with/without AI assistance does not apply.

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

Yes, the study primarily focused on standalone performance of the RIDASCREEN® Norovirus 3rd Generation EIA. The device is a qualitative enzyme immunoassay that provides a positive or negative result. The "Overall Clinical Trial Performance" table directly compares the results of the RIDASCREEN® assay against the "Clinical Diagnostic Truth" reference standard. There is no indication of human interpretation influencing the ELISA's output or any human-in-the-loop step being assessed as part of its performance.

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

The ground truth used was a composite reference standard (referred to as "Clinical Diagnostic Truth"). This was established through a combination of molecular assays (conventional RT-PCR with bi-directional sequencing for regions B and D, and quantitative real-time RT-PCR initially) and electron microscopy (EM). This is a laboratory-based, objective ground truth, which is often considered highly reliable for viral detection.

8. The sample size for the training set

The document does not explicitly mention a "training set" in the context of device development or a formal machine learning model. For this device, an immunoassay, the concept of a "training set" as understood in AI/ML is not directly applicable. Performance characteristics (like precision, detection limit, cross-reactivity, and strain reactivity) were evaluated using panels of samples, but these are typically part of analytical validation rather than training a model.

The "Strain-reactivity" section mentions the sponsor studied 100 archived samples (80 positive across different genotypes, 10 negative, 10 other viruses) but this was for analytical characterization, not model training.

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

As there's no explicitly defined "training set" in the AI/ML sense, a ground truth establishment method for it isn't described. For the analytical studies (e.g., LoD, strain reactivity) that used well-characterized samples, the ground truth was based on known viral concentrations or confirmed genotypes through methods like conventional RT-PCR followed by sequencing and electron microscopy.

Ask a specific question about this device