Search Results

BD Respiratory Viral Panel for BD MAX™ System is an automated multiplexed real-time reverse transcriptase polymerase chain reaction (RT- PCR) test intended for the simultaneous, qualitative detection and differentiation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), influenza A, influenza B, and/or respiratory syncytial virus (RSV) nucleic acid in nasopharyngeal swab (NPS) and anterior nasal swab (ANS) specimens from individuals with signs and symptoms of respiratory tract infection. Clinical signs and symptoms of respiratory tract infection due to SARS-COV-2, influenza, and RSV can be similar.

BD Respiratory Viral Panel for BD MAX™ System is intended for use as an aid in the differential diagnosis of SARS-CoV-2, influenza A, influenza B, and/or RSV infection if used in conjunction with other clinical and epidemiological information, and laboratory findings. SARS-CoV- 2, influenza A, influenza B, and RSV viral nucleic acid are generally detectable in NPS and ANS specimens during the acute phase of infection.

Positive results do not rule out co-infection with other organisms. The agent(s) detected by the BD Respiratory Viral Panel for BD MAX™ System may not be the definitive cause of disease.

Negative results do not preclude SARS-CoV-2, influenza B, and/or RSV infection.

The results of this test should not be used as the sole basis for diagnosis, treatment, or other patient management decisions.

BD Respiratory Viral Panel-SCV2 for BD MAX™ System:

BD Respiratory Viral Panel-SCV2 for BD MAX™ System is an automated multiplexed real-time reverse transcriptase polymerase chain reaction (RT-PCR) test intended for the simultaneous, qualitative detection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral nucleic acid in nasopharyngeal swab (NPS) and anterior nasal swab (ANS) specimens from individuals with signs and symptoms of respiratory tract infection. SARS-CoV-2 viral RNA is generally detectable in NPS and ANS specimens during the acute phase of infection.

The BD Respiratory Viral Panel-SCV2 for BD MAX™ System is intended for use as an aid in the diagnosis of SARS-CoV-2 infection if used in conjunction with other clinical and epidemiological information, and laboratory findings.

Positive results do not rule out co-infection with other organisms. The agent detected by the BD Respiratory Viral Panel-SCV2 for BD MAX™ System may not be the definitive cause of disease.

Negative results do not preclude SARS-CoV-2 infection.

The results of this test should not be used as the sole basis for diagnosis, treatment management decisions.

The BD Respiratory Viral Panel (BD RVP) and BD Respiratory Viral Panel-SCV2 (BD RVP-SCV2) along with the BD MAXTM System are comprised of an instrument with associated hardware and accessories, disposable microfluidic cartridges, master mixes, unitized reagent strips, and extraction reagents. The instrument automates sample preparation including target lysis, Total Nucleic Acid (TNA) extraction and concentration, reagent rehydration, target nucleic acid amplification and detection using real-time PCR. The assay includes a Sample Processing Control (SPC) that is present in the Extraction Tube. The SPC monitors RNA extraction steps, thermal cycling steps, reagent integrity and the presence of inhibitory substances. The BD MAX™ System software automatically interprets test results. For the BD Respiratory Viral Panel for BD MAX™ System and BD Respiratory Viral Panel-SCV2 for BD MAX™ System, a test result may be called as POS, NEG or UNR (Unresolved) based on the amplification status of the targets and of the Sample Processing Control. IND (Indeterminate) or INC (Incomplete) results are due to BD MAX™ System failure.

The provided text describes the analytical and clinical performance evaluation of two molecular diagnostic devices: the BD Respiratory Viral Panel (BD RVP) for BD MAX System and the BD Respiratory Viral Panel-SCV2 (BD RVP-SCV2) for BD MAX System. Both are real-time RT-PCR tests for detecting respiratory viruses in nasopharyngeal and anterior nasal swab specimens.

Here's an analysis of the acceptance criteria and study proving device performance, based on the provided text:

Important Note: The provided text is a summary of a 510(k) premarket notification. As such, it reports performance data but does not explicitly state "acceptance criteria" in the format of a pre-defined table with specific percentage goals for PPA and NPA. Instead, the reported performance itself serves as the demonstration that the device is "substantially equivalent" to predicate devices, implying that the observed performance would be considered acceptable by the FDA for the given intended use. For this response, I will interpret "acceptance criteria" as the performance metrics and results deemed sufficient for clearance.

1. Table of Acceptance Criteria (Interpreted) and Reported Device Performance

Given that this is a 510(k) submission, the "acceptance criteria" are not explicitly listed with numeric targets in the document. Instead, the reported performance is presented to demonstrate substantial equivalence to legally marketed predicate devices. The implicit acceptance criteria are the high percentages of Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) achieved by the device when compared to a composite reference method (ground truth).

The primary performance data comes from clinical evaluation studies for both the BD RVP and BD RVP-SCV2.

BD Respiratory Viral Panel (BD RVP) for BD MAX System - Clinical Performance Summary

BD Respiratory Viral Panel-SCV2 (BD RVP-SCV2) for BD MAX System - Clinical Performance Summary

For BD RVP-SCV2 clinical performance, the text refers to the same clinical study results as BD RVP, but only for SARS-CoV-2.

• SARS-CoV-2 (NP Swab): PPA 34.6% (541/1562), NPA not explicitly listed but implied from overall subject data. This seems to be a positivity rate, not an agreement rate. The clinical performance summary tables (Table 28 and 29) for SARS-CoV-2 are the relevant ones.
  • NP (Overall): PPA 98.9%, NPA 97.7%
  • ANS (Overall): PPA 98.4%, NPA 97.7%
    (These are the same as reported for the full BD RVP, which makes sense as SCV2 is a subset).

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

BD Respiratory Viral Panel (BD RVP) for BD MAX System:

• Prospective Clinical Evaluation:

  • Sample Size:
    • 2,005 subjects enrolled.
    • Test Set: 1,562 nasopharyngeal swabs (NPS) and 1,566 anterior nasal swabs (ANS).
    • For performance calculations, slightly fewer compliant specimens were used:
      • NPS: 1,545 for SARS-CoV-2, 1,562 for Flu A, Flu B, and RSV.
      • ANS: 1,561 for SARS-CoV-2, 1,564 for Flu A, Flu B, and RSV.
  • Data Provenance:
    • Country of Origin: United States (six geographically distinct U.S. study sites) and Europe (two geographically distinct sites).
    • Retrospective or Prospective: Primarily Prospective. Specimens were collected between January and August 2022.
      • January to early April 2022: Prospective archived/frozen (Category II).
      • Mid-April to August 2022: Prospective fresh (Category I).

• Retrospective Clinical Evaluation:

  • Sample Size:
    • NPS: 240 frozen retrospective nasopharyngeal swabs.
    • Nasal Swabs: 187 frozen retrospective nasal swabs.
  • Data Provenance:
    • Country of Origin: Not explicitly stated within the retrospective section, but implied to be part of the broader US/Europe context from the prospective study.
    • Retrospective or Prospective: Retrospective. Specimens collected between December 2019 and January 2022 (for NPS) and February 2021 and February 2023 (for nasal swabs) from external sources. These were "archived, frozen specimens".

BD Respiratory Viral Panel-SCV2 for BD MAX System:

• The text states: "In the BD Respiratory Viral Panel-SCV2 for BD MAX™ System clinical study, reportable results from specimens compliant at the specimen and PCR levels were obtained from 8 geographically diverse sites. Nasopharyngeal and nasal specimens totaled 1.566 respectively." This refers to the same clinical study data as the broader BD RVP, specifically focusing on SARS-CoV-2 results from that dataset.

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 experts or their qualifications for establishing ground truth.

Instead, the ground truth for the clinical evaluation (test set) was established by comparison to highly sensitive molecular assays (NAATs).

• For SARS-CoV-2: A composite method of "two out of three highly sensitive molecular assays (NAATS) that are FDA authorized under EUA for SARS-CoV-2."
• For influenza B and RSV: "an FDA-cleared high sensitivity RT-PCR assay."

This indicates an analytical or "reference method" based ground truth rather than a human expert consensus.

4. Adjudication Method for the Test Set

The adjudication method relies on a "two out of three" rule for SARS-CoV-2 and a single FDA-cleared RT-PCR assay for influenza B and RSV.

• SARS-CoV-2: "Any specimen that tested positive by two EUA assays was considered positive for SARS-CoV-2, whereas any specimen that tested negative by two EUA assays was considered negative."
• Influenza B and RSV: Comparison to a single "FDA-cleared high sensitivity RT-PCR assay."

No explicit 2+1 or 3+1 human expert adjudication method is mentioned, as the ground truth derivation is assay-based.

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

No, an MRMC comparative effectiveness study was not done. This device is a diagnostic assay (RT-PCR test), not an AI-assisted imaging product. Therefore, the concept of "human readers improve with AI vs without AI assistance" does not apply. The performance is assessed on the device's ability to accurately detect specific viral nucleic acids.

6. If a Standalone (i.e. algorithm only without human-in-the loop performance) Was Done

While the term "algorithm" is used to describe the software interpreting PCR data, the entire device (BD MAX System with the viral panel) is an automated system. The performance data presented (PPA, NPA) is the standalone performance of the device without human interpretation of the raw PCR signals for diagnosis. The BD MAX™ System software "automatically interprets test results" (page 7). Therefore, the clinical performance described is the standalone performance.

7. The Type of Ground Truth Used

The ground truth used was reference method comparison / composite molecular assay results:

• SARS-CoV-2: Composite reference method using results from "two out of three highly sensitive molecular assays (NAATs) that are FDA authorized under EUA for SARS-CoV-2." (Page 42)
• Influenza A, Influenza B, and RSV: An "FDA-cleared high sensitivity RT-PCR assay." (Page 42)

This is a form of "analytical ground truth" established by other validated diagnostic tests, rather than pathology, expert consensus on images, or long-term clinical outcomes.

8. The Sample Size for the Training Set

The document does not explicitly state a separate "training set" sample size for the assay's development in the context of a statistical machine learning model.

However, it mentions: "As the product undergoes product development, the data is supplemented, and the algorithm is adjusted ('trained') using viral cultures spiked into clinical background matrices at levels surrounding the limit of detection and expected clinical range." (Page 30, and repeated on Page 40 for SCV2). This refers to internal development and optimization using contrived samples, not a distinct, pre-defined "training set" of patient data for a typical AI/ML model for regulatory submission.

Performance data for the submission is based on the "test set" (clinical evaluation studies described in section 2) and analytical studies.

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

As noted above, there isn't a "training set" ground truth in the sense of independently verified patient data used to directly train a submitted AI model. Instead, the "algorithm" (how the BD MAX software interprets PCR signals to call positive/negative) was "adjusted ('trained')" using:

• Viral cultures spiked into clinical background matrices at various concentrations (e.g., around the Limit of Detection and expected clinical range).
• These are contrived samples with a known (positive or negative) status and precise viral concentrations.

The ground truth for these internal development/training stages would be the known concentration of spiked viral material.

Ask a specific question about this device

The DiaSorin Molecular Simplexa™ COVID-19 Direct is real-time RT-PCR assay intended for use on the LIAISON® MDX instrument for the in vitro qualitative detection of nucleic acid from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in nasopharyngeal swabs (NPS) and nasal swabs (NS) from symptomatic individuals suspected of COVID 19 by their healthcare provider. The Simplexa™ COVID-19 Direct assay is an aid in the diagnosis of SARS-CoV-2 infection.

Positive results are indicative of the presence of SARS-CoV-2 RNA. Clinical correlation with patient history and other diagnostic information is necessary to determine patient infection status. Positive results do not rule out co-infection with other pathogens. Negative results do not preclude SARS-CoV-2 infection and should not be used as tor patient management decisions.

Results are meant to be used in conjunction with other clinical, epidemiologic, and laboratory data, in accordance with the guidelines provided by the relevant public health authorities.

The Simplexa COVID-19 Direct is a real-time RT-PCR (rRT-PCR) system that enables the direct amplification and detection of SARS-CoV-2 (COVID-19) RNA from nasopharyngeal swab or nasal swab that has not undergone nucleic acid extraction. The system consists of the Simplexa COVID-19 Direct reaction mix, the LIAISON MDX (with LIAISON MDX Studio Software), the Direct Amplification Disc and associated accessories. The assay uses forward and reverse primers and associated fluorescent probe(s) included in the reaction mix to amplify SARS-CoV-2 cDNA reverse transcribed from RNA. The primers and probe sets are designed to detect SARS-CoV-2 ORF 1ab and S gene from the viral RNA in nasopharyngeal swab or nasal swab. An RNA internal control, with associated primers and a fluorescent probe, is included in the reaction mix to detect RT-PCR failure and/or inhibition.

Here's a breakdown of the acceptance criteria and study proving the device meets them, based on the provided text:

Device: Simplexa™ COVID-19 Direct

1. Table of Acceptance Criteria and Reported Device Performance

For the Simplexa™ COVID-19 Direct assay, the primary acceptance criteria revolve around its accuracy in detecting SARS-CoV-2 (COVID-19) RNA in patient samples, as well as its reproducibility, analytical sensitivity (Limit of Detection), analytical reactivity (ability to detect various strains), and specificity (cross-reactivity and interference).

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

• Clinical Agreement Test Set:

  • Total Samples: 1,150 prospective (fresh and/or frozen) samples collected.
  • Samples analyzed: 1,011 samples (114 excluded due to insufficient evidence for media types, 24 invalid results, 1 indeterminate CRM result).
  • Breakdown: 299 Nasopharyngeal Swabs (NPS) and 712 Nasal Swabs (NS).
  • Provenance: Collected from four (4) geographically diverse collection sites, one of which was outside the United States (OUS). Samples were prospective (fresh and/or frozen).
  • Timeframe: October 2020 to April 2021.

• Reproducibility Test Set:

  • Total replicates: 90 replicates per panel member (4 panel members), totaling 360 individual tests.
  • Panel members: 2 contrived low positive (LP), 2 contrived moderate positive (MP), 1 positive control, 1 negative (UTM).
  • Provenance: Tested at two (2) external clinical sites and one (1) internal site.
  • Study Design: Each panel member tested in triplicate per run, for 2 runs per day, for 5 non-consecutive testing days. Each site had two operators.

• Analytical Sensitivity (LoD) Test Set:

  • NPS: 40 replicates for confirmation.
  • NS: 20 replicates for confirmation.
  • WHO International Standard: 40 replicates for confirmation.

• Analytical Reactivity (Wet testing) Test Set:

  • 3 replicates per strain for 5 SARS-CoV-2 strains.

• Cross-Reactivity Test Set:

  • 3 replicates per organism for 47 different viruses, bacteria, and fungi (some 6 replicates for Leptospira interrogans).

• Potential Interfering Substances Test Set:

  • 3 replicates per substance (some 6 replicates for saliva and Zanamivir).

• Interference by Other Microorganisms Test Set:

  • 3 replicates per organism (some 6 replicates for Lactobacillus plantarum 17-5).

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

The ground truth for the clinical agreement test set was established using a "Composite Reference Method (CRM)" based on three (3) COVID-19 EUA approved NAAT assays. The rule for CRM agreement was: "Two out of three positive results determined 'Detected' CRM and two out of three negative results determined 'Not Detected' CRM."

The document does not specify the number or qualifications of experts (e.g., medical technologists, clinical lab scientists, or physicians) who performed these NAAT assays or interpreted their results for the CRM. It's implied that these were standard laboratory personnel qualified to run EUA-approved molecular diagnostic tests.

For analytical studies (LoD, reproducibility, reactivity, cross-reactivity, interference), the ground truth was based on the known composition and concentration of the samples (e.g., spiked RNA, cultured organisms, negative matrix). No external experts beyond the study design team would have been needed for this type of ground truth establishment.

4. Adjudication Method for the Test Set

For the clinical agreement test set, the adjudication method for the ground truth (CRM) was clearly defined: "Two out of three positive results determined 'Detected' CRM and two out of three negative results determined 'Not Detected' CRM." This is a form of consensus-based adjudication, specifically a majority rule.

For other analytical studies, adjudication was not described as it involved pre-defined positive/negative samples rather than interpretive human judgment.

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. This study focuses on the in vitro diagnostic performance of a molecular assay (RT-PCR) in a laboratory setting, not on the interpretative performance of human readers (e.g., radiologists) with or without AI assistance. Therefore, there is no discussion of human readers or an effect size of AI assistance.

6. If a Standalone (i.e. algorithm only without human-in-the loop performance) Was Done

Yes, the primary clinical performance and analytical studies are standalone (algorithm only) performance. The Simplexa™ COVID-19 Direct is an RT-PCR assay. Its "performance" refers to its ability to detect SARS-CoV-2 RNA based on its set algorithms for signal detection (Ct values for S gene and ORF1ab targets) and interpretation. The results (detected/not detected) are determined directly by the instrument and its software, not by a human interpreting images or complex patterns. The human involvement is in sample preparation and loading, and reviewing the qualitative output from the instrument.

7. The Type of Ground Truth Used

• Clinical Agreement Test Set: Composite Reference Method (CRM) using results from three (3) COVID-19 EUA approved NAAT assays, with a "two out of three" majority rule for determining "Detected" or "Not Detected." This is a form of expert consensus based on other validated diagnostic tests.
• Analytical Studies (Reproducibility, LoD, Reactivity, Cross-Reactivity, Interference): Known analytical truth established by spiking known concentrations of inactivated viral particles or other organisms into negative matrices. This is a laboratory-controlled ground truth.

8. The Sample Size for the Training Set

The provided document describes a premarket notification (510(k)) for an in vitro diagnostic device. For such devices, particularly RT-PCR assays, the "training set" typically refers to internal development and optimization data, rather than a distinct, formally defined "training set" for machine learning algorithms that would be tested on a separate "test set."

The document does not specify a numerical sample size for a training set. The assay's design (primers, probes, conditions) would have been developed and optimized internally by DiaSorin Molecular using various samples and experiments, but these are not enumerated as a specific "training set" in this regulatory submission. The "in silico inclusivity analysis" section points to the use of GISAID databases (millions of sequences) which could be considered a form of "training data" for validating the generalizability of the primer/probe design, but not as a conventional, labeled "training set" in a machine learning context.

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

Since a formal "training set" with established ground truth is not explicitly detailed in the way a machine learning model's training data would be, we can infer the following:

• Assay Development & Optimization: The ground truth for the development phase would have been based on known positive and negative samples, viral loads, and various SARS-CoV-2 strains or synthetic genetic material. This involves standard molecular biology techniques where the presence or absence of the target nucleic acid, and its concentration, are experimentally determined and controlled.
• In Silico Inclusivity: For the evaluation of primer/probe design against genetic variants, the "ground truth" is the published, annotated SARS-CoV-2 genome sequences available in the GISAID database. This involves bioinformatic analysis to determine sequence homology and potential binding efficacy.

Ask a specific question about this device

The BioFire® COVID-19 Test 2 is a qualitative nested multiplexed RT-PCR in vitro diagnostic test intended for use with the BioFire® FilmArray® 2.0 and BioFire® FilmArray® Torch Systems. The BioFire COVID-19 Test 2 detects nucleic acids from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in nasopharyngeal swabs (NPS) from symptomatic individuals suspected of COVID-19 by their healthcare provider.

Results are for the identification of SARS-CoV-2 RNA. The SARS-CoV-2 RNA is generally detectable in NPS specimens during the acute phase of infection. Positive results are indicative of the presence of SARS-CoV-2 RNA; clinical correlation with patient history and other diagnostic information is necessary to determine patient infection status. Positive results do not rule out co-infection with other pathogens.

Results are meant to be used in conjunction with other clinical, epidemiologic, and laboratory data, in accordance with the guidelines provided by the relevant public health authorities. The BioFire COVID-19 Test 2 is intended for use by trained medical and laboratory professionals in a laboratory setting or under the supervision of a trained laboratory professional.

For In Vitro Diagnostic Use.

The BioFire COVID-19 Test 2 is a multiplexed nucleic acid-based test for the detection of SARS-CoV-2 RNA from nasopharyngeal swabs (NPS) eluted in either transport medium or saline. The test was originally described and cleared in K211079. The BioFire COVID-19 Test 2 uses BioFire FilmArray technology and is for use with BioFire FilmArray 2.0 and BioFire FilmArray Torch instruments. Once the sample is injected into the FilmArray pouch is loaded into the Film Array instrument which performs all aspects of testing including nucleic acid extraction, reverse-transcription, and nested PCR with melt analysis. The currently cleared version of the test uses three SARS-CoV-2 assays and returns a 'SARS-CoV-2 Detected' call if one or more of the SARS-CoV-2 assays are positive.

The purpose of this submission is to display results for four additional SARS-CoV-2 assays which are currently present on the test, but for which results are masked through software. The assays are being unmasked as a mitigation against the risk of future SARS-CoV-2 variants affecting the sensitivity of the BioFire COVID-19 Test 2 due to mutations in assay primer regions. Note that to date BioFire Defense has not identified any variants that are predicted to affect the performance of the three-assay version of BioFire COVID-19 Test 2 described in K211079. These changes are being requested preemptively. The calling scheme when using the seven total SARS-CoV-2 assays will remain unchanged: one or more positive SARS-CoV-2 assay results will return an overall 'SARS-CoV-2 Detected' result.

Here's a breakdown of the acceptance criteria and the study details for the BioFire COVID-19 Test 2 based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The provided document describes a Special 510(k) submission where the primary change is the unmasking of four additional SARS-CoV-2 assays within an already cleared device (BioFire COVID-19 Test 2, K211079). Therefore, the "acceptance criteria" are implied by demonstrating equivalence to the predicate device's performance. The re-analysis of prior studies with the modified software is used to show this equivalence.

Since this is a submission for a software modification to unmask existing assays and not a de novo submission establishing new performance benchmarks, the acceptance criteria are largely defined by matching or showing no significant degradation from the predicate device's performance.

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

The document states that no additional testing was performed for this submission. Instead, all study data previously submitted in K211079 were re-analyzed using the updated software.

Therefore, the "test set" for this specific submission is the re-analysis of data from the original K211079 (and potentially EUA200044) studies.

• Clinical Testing (DF-SDY-030617):
  • Sample Size: 69 positive samples, 454 negative samples. (Total = 523 samples)
  • Data Provenance: Prospective Clinical Evaluation. The document does not explicitly state the country of origin, but generally, FDA submissions for predicate devices are expected to include data from diverse geographic regions within the US, or from regions with comparable patient populations.
• Bench Testing: Sample sizes vary per study. For example:
  • LoD studies (DF-SDY-029904, DF-SDY-030331) involve serial dilutions and replicates.
  • Specificity/Exclusivity (DF-SDY-029903, DF-SDY-030333) involve testing a panel of organisms/viruses.
  • Reproducibility (DF-SDY-030398) involves testing replicates across multiple sites.

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

The ground truth for the clinical samples (DF-SDY-030617) would typically be established by a reference method, most commonly another FDA-authorized RT-PCR assay. The document does not specify the number or qualifications of experts involved in determining the ground truth for the clinical samples. For molecular diagnostics, "expert consensus" is less common for ground truth than established reference tests.

For bench testing studies (e.g., LoD, inclusivity, exclusivity), the ground truth is based on the known concentration of spiked analytes or the known identity of the assayed organisms/viruses, which does not typically involve human experts establishing ground truth in the same way as clinical image interpretation.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method for the test set regarding human interpretation, as the device is an in vitro diagnostic (IVD) molecular test for direct detection of SARS-CoV-2 RNA. Results are determined by the instrument and its software.

For the clinical study, the reference method used to establish positive/negative status for the clinical samples would be the "adjudication." However, the method (e.g., comparison to a composite reference standard, or another cleared test) is not detailed here.

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

No, an MRMC comparative effectiveness study was not done. This device is an in vitro diagnostic (IVD) device that performs a laboratory test. It does not involve human readers interpreting images or data with or without AI assistance. The output is a "SARS-CoV-2 Detected" or "SARS-CoV-2 Not Detected" result.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) was done

Yes, the performance presented is standalone/algorithm only. The BioFire COVID-19 Test 2 is an IVD automated system. The results are generated directly by the device's software based on its internal processes (nucleic acid extraction, RT-PCR, melt analysis) without human interpretation of raw data. The study validates the device's ability to detect SARS-CoV-2 RNA independently. The phrasing "Software uses results from 7 assays" further confirms this.

7. The Type of Ground Truth Used

• Clinical Testing (DF-SDY-030617): The ground truth for the prospective clinical evaluation samples would most commonly be established by a highly sensitive and specific reference molecular assay (e.g., another FDA-cleared or EUA RT-PCR test for SARS-CoV-2), possibly combined with clinical diagnosis, but the document does not explicitly state this.
• Bench Testing:
  • LoD, Reactivity (Inclusivity): Ground truth is based on the known concentration and identity of specific SARS-CoV-2 strains/genomic material spiked into negative matrix.
  • Specificity (Exclusivity): Ground truth is based on the known identity of other respiratory pathogens or commensals tested.
  • Interfering Substances: Ground truth is based on the known presence of potential interfering substances without SARS-CoV-2.
• In Silico Analysis: Ground truth is based on publicly available genetic sequence databases (e.g., GISAID for SARS-CoV-2 variants).

8. The Sample Size for the Training Set

The document does not mention a separate training set for this submission. The purpose of this submission is to unmask existing assays on an already established device. The performance data presented are from validation and verification studies (effectively "test sets") previously conducted for the predicate device. For the original development of the BioFire COVID-19 Test 2 assays, various internal optimization and calibration steps (which could be considered analogous to "training") would have occurred, but these details are not part of this 510(k) summary.

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

As no specific "training set" is described for this 510(k) modification, this question is not directly applicable. For the initial development of the assays, the ground truth would have been established through a combination of:

• Bioinformatics: Designing primers and probes based on known SARS-CoV-2 sequences.
• Analytical studies: Testing synthetic targets and cultured virus at known concentrations.
• Clinical studies: Initial evaluation with patient samples where SARS-CoV-2 status was determined by a reference method.

Ask a specific question about this device