Search Results

The DPP Zika IgM System is intended for the presumptive qualitative detection of Zika virus IgM antibodies in human serum (plain or separation gel), potassium-EDTA plasma, potassium-EDTA venous whole blood, or fingerstick whole blood specimens, collected from individuals meeting the CDC Zika virus clinical criteria (e.g., a history of clinical signs and symptoms associated with Zika virus infection) and/or CDC Zika virus epidemiological criteria (e.g., history of residence in or travel to a geographic region with active Zika transmission at the time of travel, or other epidemiological criteria for which Zika virus testing may be indicated). Specimens from symptomatic patients or returning travelers from endemic areas should not be collected prior to 8 days after symptom onset or after potential exposure as a sample collected earlier may return a negative result. If testing is needed after day 8 and results are negative, testing must be repeated one week later. Positive results must be confirmed by following the latest CDC guidelines for the diagnosis of Zika virus infection.

Results of this test are intended to be used in conjunction with clinical observations, patient history, epidemiological information, and other laboratory results. Zika IgM levels over the course of illness are not well characterized. Zika IgM levels are variable during the course of infection and may be detectable near day 4 post-onset of symptoms and persist up to approximately 12 weeks following initial infection.

Negative results may be seen in specimens collected before day four post-onset of symptoms or after the window of detectable IgM closes and therefore do not preclude the possibility of Zika virus infection, past or present.

The Chembio DPP Zika IgM System is not indicated for testing blood or plasma donors.

The test cannot be visually interpreted by the operator and must be read on the DPP Micro Reader.

DPP Zika IgM System Control Pack

The Chembio DPP Zika IgM System Control Pack is an external quality control kit for use with the DPP Zika IgM System only. The performance characteristics of the DPP Zika IgM System Control Pack have not been established for any other assay or instrument different from the DPP Micro Reader.

DPP Micro Reader

The DPP Micro Reader is a reflectance reader used to obtain test results from DPP Zika IgM System. The DPP Micro Reader is necessary to minimize errors from direct visual interpretation; the results of DPP Zika IgM System cartridges must be read exclusively with the DPP Micro Reader.

Chembio' s DPP Zika IgM System is a qualitative immunochromatographic assay for the presumptive detection of IgM antibodies to Zika virus. The DPP Zika IgM System is composed of:

• A single-use immunochromatographic test for the presumptive detection of ZIK V 1. IgM antibodies in human serum (plain or separation gel), potassium-EDTA plasma, potassium-EDTA venous whole blood, or fingerstick whole blood specimens.

• 2. The DPP Micro Reader to minimize errors from direct visual interpretation.

The document describes the Chembio DPP Zika IgM System, DPP Zika IgM System Control Pack, and DPP Micro Reader. The system is intended for the presumptive qualitative detection of Zika virus IgM antibodies in human serum, plasma, and whole blood specimens.

Here's a breakdown of the acceptance criteria and study details:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the DPP Zika IgM System are primarily demonstrated through its Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) compared to a predicate EUA or FDA-cleared comparator assay, across various sample types. Analytical performance (precision, cross-reactivity, interference, analytical sensitivity, and stability) also forms part of the acceptance.

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

• Serum:
  • Positive Predictive Agreement:
    • 99 samples from symptomatic individuals in Peru (retrospective).
    • 11 samples from individuals in the Dominican Republic (retrospective).
    • 32 samples from 26 individuals in Brazil during arbovirus outbreaks (retrospective).
    • FDA Plasma Zika Panel: 24 Zika IgM positive, 12 negative.
  • Negative Predictive Agreement (Endemic Area): 250 subjects from a Zika endemic area (prospectively collected).
  • Negative Predictive Agreement (Non-Endemic Area): 250 subjects from a Zika non-endemic area (prospectively collected).
• Potassium-EDTA Plasma:
  • Positive Predictive Agreement (Manufacturer Study): 299 IgM antibody samples from 48 individuals (from 50 symptomatic subjects) in the Dominican Republic (archived, confirmed by NAT). Includes 12 pregnant women.
  • Positive Predictive Agreement (External Sites): 171 comparator positive IgM antibody samples from 39 individuals in the Dominican Republic (archived). Also, 49 prospectively collected asymptomatic pregnant women from the continental United States (negative by comparator assay) were interspersed.
  • Negative Predictive Agreement (Endemic Area): 250 subjects from a Zika endemic area (prospectively collected).
  • Negative Predictive Agreement (Non-Endemic Area): 250 subjects from a Zika non-endemic area (prospectively collected).
• Potassium-EDTA Venous Whole Blood:
  • Positive Predictive Agreement (Internal/External Sites): 41 plasma samples (from a plasma replacement study) from 6 individuals in the Dominican Republic; 10 frozen natural whole blood samples from individuals in the Dominican Republic; 171 antibody positive plasma specimens (from plasma replacement study) plus 49 antibody negative specimens from asymptomatic pregnant women from the US.
  • Positive Predictive Agreement (Contrived): 299 subjects for analysis (from 300 "all comers" at 3 US clinics); samples were contrived by potassium-EDTA plasma replacement.
  • Negative Predictive Agreement (Endemic Area): 250 subjects from a Zika endemic area (prospectively collected).
  • Negative Predictive Agreement (Non-Endemic Area): 250 subjects from a Zika non-endemic area (prospectively collected).
• Fingerstick Whole Blood:
  • Positive Predictive Agreement (Contrived): 372 subjects for analysis (from 375 adult subjects across 4 US near-patient sites, "all comers" basis); samples were pre-spiked to create contrived samples.
  • Negative Predictive Agreement (Endemic Area): 250 subjects from a Zika endemic area (prospectively collected).
  • Negative Predictive Agreement (Non-Endemic Area): 250 subjects from a Zika non-endemic area (prospectively collected).
• Cross-Reactivity: 329 specimens for various organisms/conditions.
• Interference: Low reactive (n=3) and normal human plasma samples (n=3) for each interfering substance.
• Analytical Sensitivity: Dilution series of WHO 1st International standard.
• Assay Cut-off: 569 natural serum samples (US, Mexico); 184 natural plasma samples (US, Peru); 215 natural venous whole blood samples (US); 102 natural capillary whole blood samples (US).
• Precision/Reproducibility: A blinded panel of four plasma samples (negative, high negative, low positive, moderate positive) tested with 3 lots of the system.
• DPP Zika IgM Control Kit Precision: Moderate positive, low positive, and negative control samples.

Data Provenance: Studies include samples from Peru, Dominican Republic, Brazil, and the United States. Many samples are retrospective (archived, historical data), while some are prospectively collected (e.g., negative predictive agreement studies).

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

The document does not explicitly state the "number of experts" or their specific "qualifications" used to establish ground truth in the traditional sense of multiple human readers independently adjudicating images/cases.

Instead, the ground truth for clinical performance studies (PPA and NPA) was established by:

• Comparison with an EUA (Emergency Use Authorization) or FDA-cleared comparator assay.
• In some cases, samples were confirmed positive by RT-PCR assay for Zika virus (e.g., Peruvian and Brazilian serum samples).
• Consensus results were provided for the FDA Plasma Zika Panel, and these are stated to be the responsibility of the FDA.
• For contrived samples (venous whole blood, fingerstick whole blood), the "ground truth" was based on expected results from spiking with known positive or negative material, corroborated by comparator testing.

4. Adjudication Method for the Test Set

The primary method for "adjudication" (or reference standard determination) appears to be comparison with another laboratory assay (EUA or FDA-cleared comparator assay). For Negative Percent Agreement, results were compared against an FDA Cleared Comparator and an Additional EUA Test, and "Both FDA Cleared and EUA tests" where results from both were in agreement. There is no mention of a traditional expert consensus or adjudication panel (e.g., 2+1, 3+1) for clinical performance in this document.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. The study focuses on the standalone performance of the DPP Zika IgM System against reference methods/comparator assays, not on how human readers improve with or without AI assistance. The DPP Micro Reader is an integral part of the device, reading results from the immunochromatographic assay, and is not an AI tool assisting human interpretation. The document explicitly states the test "cannot be visually interpreted by the operator and must be read on the DPP Micro Reader," meaning it's not a human-in-the-loop scenario.

6. If a Standalone Study Was Done

Yes, a standalone study was done. The document provides extensive data on the performance of the DPP Zika IgM System (algorithm only, as it's an automated reader) in detecting Zika virus IgM antibodies across various sample types and clinical scenarios, independently compared to established reference standards (comparator assays, RT-PCR, or expected reactivity from spiked samples).

7. The Type of Ground Truth Used

The ground truth used various forms:

• Comparator Assay Results: Results from an EUA or FDA-cleared Zika IgM comparator assay.
• RT-PCR Confirmation: For some positive samples, PCR confirmation for Zika virus was used.
• Consensus Data: For the FDA Plasma Zika Panel, consensus results were utilized.
• Expected Reactivity from Spiked Samples: For contrived samples (venous whole blood, fingerstick whole blood), the ground truth was based on the known positive or negative status of the spiked material.

8. The Sample Size for the Training Set

The document does not explicitly specify a "training set" in the context of machine learning. The DPP Micro Reader and assay are likely developed based on extensive R&D and optimization, but the provided performance data relates to validation studies, not an explicit training set for a distinct AI model. The product is an in-vitro diagnostic device that produces quantitative results read by a device, not a machine learning algorithm requiring a separate training dataset in the typical sense.

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

Since an explicit "training set" for a machine learning model is not described, the method for establishing its ground truth is not applicable or detailed in this document. The development of the assay and the Micro Reader's algorithm would have involved internal validation and optimization using various samples, but these are part of product development rather than a defined, separate "training set" for regulatory evaluation in the context of this 510(k) summary.

Ask a specific question about this device

The DiaSorin LIAISON® XL Zika Capture IgM II assay is intended for the presumptive qualitative detection of Zika virus IgM antibodies in human sera collected from individuals meeting CDC Zika virus clinical criteria (e.g., a history of clinical signs and symptoms associated with Zika virus infection) and/or CDC Zika virus epidemiological criteria (e.g., history of residence in or travel to a geographic region with active Zika transmission at the time of travel, or other epidemiological criteria for which Zika virus testing may be indicated). Specimens from symptomatic patients or returning travelers from endemic areas must be collected not earlier than day 8 after the onset of symptoms or risk of exposure, respectively. Positive results must be confirmed by following the latest CDC quidelines for the diagnosis of Zika virus infection.

Results of this test are intended to be used in combination with clinical observations, patient history, epidemiological information, and other laboratory evidences. Zika IqM levels over the course of illness are not well characterized. IqM levels are variable, may be detectable near day 4 post onset of symptoms and persist up to approximately 12 weeks following initial infection.

Negative results may be seen in specimens collected before day four post onset of symptoms or after the window of detectable IgM closes, and therefore do not preclude the possibility of Zika virus infection, past or present.

This LIAISON® XL Zika Capture IgM II assay is not indicated for testing blood or plasma donors.

The test has to be performed on the LIAISON® XL Analyzer.

The DiaSorin LIAISON® XL Zika Capture IgM II Control Set is intended for use as assayed quality control samples to monitor the performance of the LIAISON® XL Zika Capture IgM II assay. The performance characteristics of the LIAISON® XL Zika Capture IgM II controls have not been established for any other assay or instrument platforms different from the LIAISON® XL.

The LIAISON® XL Zika Capture IqM II assay is an automated immunoassay utilizing chemiluminescent (CLIA) detection technology for the detection of human IgM antibodies against Zika Virus in patient sera.

The LIAISON® XL Zika Capture IgM II assay consists of the components described in the following tables.

Magnetic Particles (2.4 mL): Magnetic particles coated with a mouse monoclonal antibody to human IgM diluted in phosphate buffer containing BSA, surfactant, and

The provided text describes the performance characteristics of the DiaSorin LIAISON® XL Zika Capture IgM II assay. However, it does not describe a study comparing the device's performance against human readers or establishing acceptance criteria in the context of an Artificial Intelligence (AI) enabled device. This document is a 510(k) summary for an in-vitro diagnostic (IVD) device, which typically focuses on demonstrating substantial equivalence to a predicate device through analytical and clinical performance studies, not AI-assisted human reading performance.

Therefore, many of the requested points related to AI acceptance criteria, MRMC studies, expert adjudication, and training/test set ground truth establishment for an AI model cannot be extracted from this document.

Here's what can be extracted, interpreted, and what is missing based on the provided text:

Device: LIAISON® XL Zika Capture IgM II Assay

Device Type: In-vitro diagnostic (IVD) immunoassay for the presumptive qualitative detection of Zika virus IgM antibodies. This is not an AI-enabled device.

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

Since this is an IVD device, the "acceptance criteria" are typically defined by performance metrics like positive percent agreement (PPA), negative percent agreement (NPA), precision (reproducibility), analytical specificity (cross-reactivity), and stability. The performance is compared against a comparator assay (predicate device) and clinical observations/nucleic acid testing for ground truth.

• Day 8-14: 94.6%
• Day 15-28: 100%
• Day 29-42: 100%
• Day 43-56: 82.4%
• Day 57-70: 77.8%
• Day 71-84: 100%
  Note: Time frame 0-7 days (35.3%) is outside claimed window of detection. Some samples in the 0-7 day window were only positive by nucleic acid testing, or negative by comparator assay but positive by the device, making direct comparison complex for this early phase. |
  | Negative Agreement (NPA)| High agreement with confirmed negative samples/comparator. | Overall: 98.6% (493/500) with comparator assay (95%CI 97.1%-99.3%)
• Non-endemic (U.S.): 99.6% (249/250)
• Endemic (Dominican Republic): 97.6% (244/250) |
  | Percision/Reproducibility | Low %CV for various samples across runs, days, lots, and sites. | Within-laboratory Precision (ZIKV-M):
• Neg Ctrl: 15.0% CV
• Pos Ctrl: 5.4% CV
• Sample #1 (low signal): 15.5% CV
• Sample #2: 11.9% CV
• Sample #3: 10.6% CV
• Sample #4: 9.5% CV
  Between-Site Reproducibility (ZIKV-M):
• Neg Ctrl: 13.8% CV
• Pos Ctrl: 9.7% CV
• Sample #1 (low signal): 15.3% CV
• Sample #2: 11.1% CV
• Sample #3: 7.0% CV
• Sample #4: 8.4% CV
  Similar comprehensive data provided for ZIKV-C with varying %CVs. |
  | Analytical Specificity/Cross-Reactivity | Low reactivity (ideally 0%) to related pathogens or interfering substances. | Cross-Reactivity:
• 1/43 (2.33%) Dengue IgM samples reactive (also reactive with comparator).
• 1/14 (7.14%) Parvovirus B19 IgM reactive (negative with comparator).
• 1/16 (6.25%) Rheumatoid Factor reactive (negative with comparator).
• No reactivity for many other tested organisms (Chikungunya, Cytomegalovirus, Epstein Barr, Varicella Zoster, Yellow Fever, West Nile, Malaria, Adenovirus, Enterovirus, Hepatitis (C/B), HSV-1/2, Rubella, Borrelia, Treponema pallidum, HAMA, ANA).
  Interfering Substances: Hemoglobin (positive interference at 10 mg/mL, minimal at 2 mg/mL). Others (Bilirubin, Triglycerides, Cholesterol, Albumin, HAMA, RF) showed no significant interference. |

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

• Positive Agreement Study:

  • Sample Size: 244 specimens from 79 subjects.
  • Data Provenance: Prospectively collected from the Dominican Republic (endemic region), including 22 pregnant women.
  • Nature: Serial serum samples from 46 symptomatic subjects, and 33 single bleeds.

• Negative Agreement Study:

  • Sample Size: 500 serum samples.
  • Data Provenance:
    • 250 subjects from a non-endemic region (continental United States - Texas and Florida, collected June 2017). Pregnancy status unknown for US subjects.
    • 250 subjects from an endemic region (Dominican Republic), including 37 pregnant women.
  • Nature: Confirmed negative for Zika IgM by a comparator assay.

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)

This section is not applicable as the ground truth was established by:

• A "commercially available Zika IgM assay" (predicate device).
• Nucleic acid testing (for positive cases).
• Clinical criteria, patient history, and epidemiological information.
  This is a diagnostic kit, not an AI review system, so human expert interpretation (like radiologists) for ground truth is not relevant in this context.

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

Not applicable. The ground truth for this IVD device's clinical performance was established by laboratory methods (comparator assay, nucleic acid tests) and patient epidemiology/clinical data, not by human expert review that would require adjudication.

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 conducted. This is an in-vitro diagnostic (IVD) test kit, not an AI-assisted diagnostic tool for human readers.

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

This refers to the device itself. The LIAISON® XL Zika Capture IgM II assay is a standalone automated immunoassay. Its performance characteristics (accuracy, precision, etc.) are reported independently of human interpretation in the way an AI algorithm's standalone performance might be. It generates a quantitative "Index value" which is then interpreted as positive, negative, presumptively positive, or presumptively recent based on pre-defined cut-off values.

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

The ground truth for the clinical studies was primarily established by:

• Comparator Assay: A "commercially available Zika IgM assay" (the predicate device)
• Nucleic Acid Testing: For confirmation of Zika virus infection in symptomatic subjects.
• Clinical/Epidemiological Data: CDC Zika virus clinical criteria (signs/symptoms) and/or epidemiological criteria (residence/travel to endemic region).

8. The sample size for the training set

The document does not describe a training set as this is a traditional IVD device, not a machine learning/AI model. The "training" of such a device is typically through its chemical and biological formulation and optimization during development, not data-driven model training.

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

Not applicable as there is no training set mentioned for an AI model.

Ask a specific question about this device

The ADVIA Centaur® Zika test is for in vitro diagnostic use in the qualitative detection of IgM antibodies to the Zika virus in human serum and plasma (potassium EDTA or lithium heparin) specimens using the ADVIA Centaur XP and ADVIA Centaur XPT systems.

The ADVIA Centaur Zika test is intended for the presumptive clinical laboratory diagnosis of Zika virus infection. The test is intended for use only in individuals (children, adolescents and adults, including pregnant women) with clinical signs and symptoms consistent with Zika virus infection, and/or meeting the CDC Zika virus epidemiological criteria (history of residence in or travel to a geographic region with active Zika transmission at the time of travel, or other epidemiological criteria for which Zika virus testing may be indicated). Positive results must be confirmed by following the latest CDC guidelines for the diagnosis of Zika virus infection.

Results of this test are intended to be used in conjunction with clinical observations, patient history, epidemiological information, and other laboratory evidence to make patient management decisions. Zika IgM levels are variable over the course of the infection, and may be detectable near day 4 post onset of symptoms and persist up to approximately 12 weeks following initial infection.

Negative results may be seen in specimens collected before day 4 post onset of symptoms or after the window of detectable IgM closes, and therefore do not preclude the possibility of Zika virus infection, past or present.

The ADVIA Centaur Zika test is not indicated for testing blood or plasma donors.

The ADVIA Centaur Zika test consists of the components described in the following table: ADVIA Centaur Zika Ab Primary Reagent ReadyPack (included in Zika Ab assay kit), ADVIA Centaur Zika Ab Lite Reagent, ADVIA Centaur Zika Ab Solid Phase Reagent, ADVIA Centaur Zika Ab Ancillary Well Reagent, ADVIA Centaur Zika Ab Calibrators (included in Zika Ab assay kit), ADVIA Centaur Zika Ab High Calibrator, ADVIA Centaur Zika Ab Low Calibrator, ADVIA Centaur Zika Ab Controls (included in Zika Ab QC kit), ADVIA Centaur Zika Ab Negative Control, ADVIA Centaur Zika Ab Low Calibrator, ADVIA Centaur Zika IgM Primary Reagent ReadyPack (included in Zika IgM assay kit), ADVIA Centaur Zika IgM Lite Reagent, ADVIA Centaur Zika IgM Solid Phase Reagent, ADVIA Centaur Zika IgM Ancillary Well Reagent, ADVIA Centaur Zika IgM Calibrators (included in Zika IgM assay kit), ADVIA Centaur Zika IgM High Calibrator, ADVIA Centaur Zika IgM Low Calibrator, ADVIA Centaur Zika IgM Controls (included in Zika IgM QC kit), ADVIA Centaur Zika IgM Negative Control, ADVIA Centaur Zika IgM Low Calibrator. The methodology is an Antibody capture immunoassay using chemiluminescence detection.

Here's an analysis of the provided text, focusing on acceptance criteria and study details for the ADVIA Centaur Zika test:

The document describes the Siemens Healthcare Diagnostics Inc. ADVIA Centaur Zika test, a qualitative assay for IgM antibodies to the Zika virus.

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly present a table of "acceptance criteria" in the format of pass/fail thresholds. However, it presents various performance metrics which can be interpreted as the data used to demonstrate the device's acceptable performance. For clarity, I've extracted the key performance metrics and values from the "Performance Characteristics" section that would typically be evaluated against pre-defined acceptance criteria.

Note: The document itself does not state specific numerical acceptance criteria (e.g., "PPA must be >X%"). Instead, it presents the achieved performance. For biological assays, "acceptance" often implies demonstrating performance comparable to or superior to current standards/predicate devices, and meeting regulatory expectations for diagnostic accuracy.

Study Details:

Most of the specified details (training set, experts for ground truth, adjudication method, MRMC study, standalone performance) are typically not included in a 510(k) summary for in vitro diagnostic (IVD) devices like this, which primarily focus on analytical and clinical performance compared to a predicate or established standard. This document is for a serological reagent, not an AI/imaging device. However, I will extract what is available from the text.

2. Sample sizes used for the test set and data provenance:

• Analytical Specificity / Cross-Reactivity:
  • 341 samples from various disease states (e.g., ANA, Dengue, Malaria, Yellow Fever Immunization). Provenance not specified beyond "specimens containing IgM antibodies against other flavivirus specimens and disease state specimens."
• FDA Zika Performance Panel:
  • Test Set (Zika IgM Consensus Positive): 24 samples
  • Test Set (Zika IgM Consensus Negative): 12 samples
  • Test Set (West Nile Virus): 11 samples
  • Test Set (Dengue Virus): 10 samples
  • Data Provenance: Samples provided by the FDA, sourced from the Blood Systems Research Institute (BSRI, now Vitalant Research Institute) from a study supported by National Institutes of Health. It appears to be retrospective due to "established consensus of sero-status."
• Clinical Studies (Zika-Positive Populations - Seroconversion Sensitivity):
  • 8 serial draws from 36 Zika PCR-positive patients.
  • Data Provenance: Dominican Republic. Prospectively collected.
• Clinical Studies (Zika-Positive Populations - Agreement):
  • Single draws from 49 Zika PCR-positive patients.
  • Data Provenance: Dominican Republic and mainland U.S. Prospectively collected.
• Clinical Studies (Zika-Negative Populations - Endemic Area):
  • Residents with symptoms, PCR-negative for Zika: 46 samples
  • Asymptomatic residents: 262 samples
  • Travelers: 47 samples
  • Total Endemic: 355 samples
  • Data Provenance: Dominican Republic, Honduras, and Puerto Rico.
• Clinical Studies (Zika-Negative Populations - Non-Endemic Area):
  • Apparently healthy male and female blood donors: 1365 samples
  • Pregnant females: 485 samples
  • Pediatric subjects (2-21 years): 128 samples
  • Total Non-Endemic: 1978 samples
  • Data Provenance: Mainland U.S.

3. Number of experts used to establish the ground truth for the test set and their qualifications:

• For the FDA Zika Performance Panel: The text states, "Performance was assessed from the subset of panel members for which an established consensus of sero-status was established." It also mentions, "The panel composition and consensus results are the responsibility of the FDA." However, the exact number and qualifications of experts contributing to this "consensus of sero-status" are not specified in this document.
• For Clinical Studies (Zika-Positive): Ground truth was established using Zika PCR-positive status. This is a direct molecular detection, not typically subject to expert consensus interpretation in the same way as imaging.
• For Clinical Studies (Zika-Negative): Ground truth was established by being "negative by the ZIKV Detect 2.0 IgM Capture ELISA" (the predicate device) and/or PCR-negative. Again, this relies on laboratory test results rather than expert interpretation of complex data.

4. Adjudication method for the test set:

• Not applicable or Not specified in the context of expert adjudication for an IVD device like this. The "consensus of sero-status" for the FDA panel implies some form of agreement, but the process is not detailed. For the clinical studies, gold standard laboratory methods (PCR, predicate device results) served as the comparator.

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. This is an in vitro diagnostic (IVD) device, not an AI or imaging device that involves human readers interpreting results. Therefore, an MRMC study is not relevant here. The device automatically generates qualitative results (reactive/non-reactive).

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

• Yes, implicitly. The ADVIA Centaur Zika test is an automated, qualitative immunoassay. Its performance is entirely standalone; it produces a result (Presumptive Zika Positive or Negative) based on its biochemical reactions and internal algorithm without human interpretation influencing the test result itself. Human involvement is in collecting samples, running the instrument, and interpreting the output for clinical context, but not in establishing the diagnostic result.

7. The type of ground truth used:

• FDA Panel: "Established consensus of sero-status" and "Zika IgM Consensus Result." This implies a combination of serological and potentially molecular data adjudicated to determine true positive/negative status for the panel samples.
• Clinical Studies (Zika-Positive): Primarily Zika PCR-positive status for initial classification.
• Clinical Studies (Zika-Negative): Primarily negative by the ZIKV Detect 2.0 IgM Capture ELISA (predicate device) and in some cases, PCR-negative.

8. The sample size for the training set:

• Not specified as a distinct "training set" in the context of device development (e.g., machine learning). For IVD assays, development involves optimizing reagents and protocols. The "Analytical Sensitivity / Assay Cut-off" section mentions using a WHO International standard for anti-Asian lineage Zika virus antibody to determine cut-off values, which is a form of calibration/optimization rather than training a model. There's no separate section describing a distinct training dataset for an algorithm.

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

• Since a "training set" in the machine learning sense is not explicitly identified, the method for establishing its ground truth is not detailed. However, for the analytical sensitivity, the ground truth was based on the WHO 1st International standard for anti-Asian lineage Zika virus antibody (human) (NIBSC 16/352), which is a highly characterized reference material with an assigned concentration.

Ask a specific question about this device

The ZIKV Detect 2.0 IgM Capture ELISA is intended for the qualitative detection of Zika virus IgM antibodies in human sera for the presumptive clinical laboratory diagnosis of Zika virus infection. The assay is intended for use only in patients with clinical signs and symptoms consistent with Zika virus infection, and/or CDC Zika virus epidemiological criteria (e.g., history of residence in or travel to a geographic region with active Zika transmission at the time of travel, or other epidemiological criteria for which Zika virus testing may be indicated). Assay results are for the presumptive detection of IgM antibodies to Zika virus (ZIKV). Positive results must be confirmed by following the latest CDC guidelines for the diagnosis of Zika virus infection.

Results of this test are intended to be used in conjunction with clinical observations. patient history, epidemiological information, and other laboratory evidence to make patient management decisions. Zika IgM levels are variable over the course of the infection, and may be detectable near day four post onset of symptoms and persist up to approximately 12 weeks following initial infection.

Negative results may be seen in specimens collected before day four post onset of symptoms or after the window of detectable IgM closes, and therefore do not preclude the possibility of Zika virus infection, past or present.

This assay is not indicated for testing blood or plasma donors.

The ZIKV Detect IgM Capture ELISA is a sandwich-type immunoassay. The test kit includes microtiter wells coated with anti-human IgM antibodies, ZIKV IgM Negative, and IgM Positive controls, ZIKV Sample Dilution Buffer, ZIKV Recombinant Antigen (Zika Ag) for IgM, Cross-reactive Control Antigen (CCA) for ZIKV IgM and normal cell antigens (NCA), secondary antibodies targeting the flavivirus antigens. The test kit also contains a HRPlabeled ZIK V-specific monoclonal antibody and tetramethylbenzidine (TMB) substrate which are used to detect ZIKV IgM antibodies in the wells.

The ZIKV Detect 2.0 IgM Capture ELISA contains sufficient reagents for one plate of 96 wells (12 x 8 strips) for human IgM targeting Zika virus. This is sufficient for testing a maximum of 28 unknown samples for human IgM, with controls included in duplicate.

Here's a breakdown of the acceptance criteria and the study details for the ZIKV Detect 2.0 IgM Capture ELISA, based on the provided text:

1. Acceptance Criteria and Reported Device Performance

The provided text doesn't explicitly state "acceptance criteria" in a separate section with specific numerical targets like "Sensitivity > X%" or "Specificity > Y%." However, from the "Performance Characteristics" and "Clinical Studies" sections, we can infer the expected performance and compare it with the reported results. The FDA's conclusion that the device's information is "sufficient" to classify it implies that the presented performance met their internal, unstated acceptance criteria for a class II device with special controls.

Here's an inferred table based on the provided results:

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

• Analytical Reproducibility Study:

  • Test Set Sample Size: 5 panel members (negative, high negative, low positive, moderate positive, re-test specimen). Each panel member had 270 replicates (3 replicates x 3 sites x 2 operators x 5 days x 3 kit lots / 3 kit lots). So, essentially 5 samples were tested extensively.
  • Data Provenance: Not explicitly stated, but implied to be laboratory-generated samples or well-characterized clinical samples for analytical studies.

• Cross-Reactivity Studies (Table 2):

  • Test Set Sample Size: 346 IgM positive serum specimens (across various microorganisms like Dengue, WNV, JE, Chikungunya, Malaria, Syphilis, etc.).
  • Data Provenance: Sourced from patients with confirmed IgM antibodies to potentially cross-reactive microorganisms. Some specific details: Yellow Fever Vaccine recipients from Colombia during a 2016 Zika outbreak; convalescent phase samples for Dengue. Seems retrospective collection.

• Interfering Substances Study:

  • Test Set Sample Size: 3 low reactive human serum samples and 3 normal human serum samples per interfering substance.
  • Data Provenance: Laboratory spiked samples.

• Analytical Sensitivity (LOD) Study:

  • Test Set Sample Size: Multiple dilutions of the WHO 1st International Standard for anti-Asian lineage Zika virus antibody (human) were tested in 20 replicates each.
  • Data Provenance: WHO International Standard, a reference material.

• Clinical Studies (Endemic and Non-Endemic Subjects):

  • Test Set Sample Size: 807 unique samples were collected from 609 subjects.
    • Endemic subjects: 353 samples.
    • Non-endemic subjects: 256 samples.
    • 744 samples had known Days Post Symptom Onset (PSO).
  • Data Provenance: Collected from endemic sites (presumed positive and presumed negative) and non-endemic sites (presumed negative). Some subjects provided serial draws up to 84 days PSO. Some from Zika endemic areas provided paired acute/convalescent draws. This is a prospective collection for the vast majority, as samples were "collected from these subjects." Geographical origin of "endemic" and "non-endemic" sites is not explicitly mentioned by country, but they are clearly differentiated.

• FDA Performance Panel:

  • Test Set Sample Size: Not explicitly stated for the entire panel, but reported for the subsets analyzed: 24 Zika IgM Consensus Positive, 12 Negative, 10 West Nile positive, 10 Dengue positive.
  • Data Provenance: Provided by the FDA from individuals infected with Zika, West Nile, or Dengue viruses at various stages of infection. This appears to be a retrospective collection of well-characterized samples.

3. Number of Experts Used to Establish Ground Truth and Qualifications

The text does not specify the number or qualifications of experts used to establish the ground truth for any of the studies.

• Clinical Studies (Endemic and Non-Endemic): The ground truth was established by a "composite reference method that included a validated Zika RT-PCR and CDC Zika MAC-ELISA." While these are laboratory tests, they would have been interpreted by trained laboratory professionals, but no specific expert count or qualification is given.
• FDA Performance Panel: Ground truth ("established consensus of sero-status") was established by the FDA. Again, specific expert details are omitted.

4. Adjudication Method for the Test Set

The text does not explicitly describe an adjudication method (like 2+1, 3+1) for resolving discrepancies in ground truth determination for any of the studies.

• Clinical Studies: The "composite reference method" combining RT-PCR and CDC Zika MAC-ELISA implies a sequential or parallel testing strategy where results from both contribute to the final ground truth. It does not mention an expert adjudication process if these reference methods yielded conflicting results.
• FDA Performance Panel: The "established consensus of sero-status" suggests that a consensus process was used by the FDA, but the details of this process (e.g., how many readers/experts, how discrepancies were resolved) are not provided.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not done. This device is an ELISA assay, which is a laboratory diagnostic tool, not an AI-assisted diagnostic imaging or interpretation system. Therefore, the concept of human readers improving with AI assistance does not apply here.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance)

Yes, the performance presented in the "Analytical performance" and "Clinical studies" sections primarily reflects the standalone performance of the ZIKV Detect 2.0 IgM Capture ELISA device. The results (e.g., PPA, NPA, cross-reactivity tables) are direct outputs of the assay based on its internal algorithms and interpretation rules described (Zika ISR, CCA/NCA ratio). While human operators perform the test and interpret the final result based on the device's guidelines, the performance metrics reported are for the device itself against the established ground truth, not for human interpretation aided by the device.

7. Type of Ground Truth Used

• Analytical Sensitivity (LOD): World Health Organization (WHO) 1st International Standard for anti-Asian lineage Zika virus antibody (human) – a standardized reference material.
• Clinical Studies (Endemic and Non-Endemic): A composite reference method consisting of a validated Zika RT-PCR and CDC Zika MAC-ELISA. This is a combination of molecular (RT-PCR) and serological (MAC-ELISA) laboratory tests.
• FDA Performance Panel: "Established consensus of sero-status" – this implies a consensus derived from laboratory tests (Zika IgM, West Nile, Dengue).

8. Sample Size for the Training Set

The document does not explicitly mention a separate "training set" for the device's development in the context of machine learning or AI models. Given that this is an ELISA assay, the concept of a training set as understood in AI/ML is not directly applicable.

However, the "Assay cut-off" section mentions:

• "The study included a testing of eight hundred thirty five (835) specimens that included 72 Zika positive, 198 other flavivirus positive, and 565 other disease positive and negative serum specimens. Receiver Operating Characteristic (ROC) curve analyses were performed to optimize for those cut-off values that maximize both sensitivity and specificity. A rudimentary bootstrap method was applied to minimize bias by any potential outliers in the sample set."
  This process of optimizing cut-off values using 835 specimens could be considered analogous to a "training" or "calibration" phase for the device's interpretation algorithm, where the rules for classifying samples (e.g., Zika Ag OD450 > Threshold, Zika ISR > 1.90) were established.

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

For the 835 specimens used to establish the assay cut-off:

• Ground Truth Establishment: The specimens were categorized as "Zika positive," "other flavivirus positive," and "other disease positive and negative." The method by which these 835 specimens were initially classified (i.e., their ground truth) is not explicitly detailed in the document. It is implied these were diagnostically confirmed samples (e.g., using reference assays like those used in the clinical studies), but the specific process is not described.

Ask a specific question about this device