Search Results

The Quidel Molecular Influenza A+B Assay is a multiplex Real Time RT-PCR assay for the in vitro qualitative detection and differentiation of influenza A and influenza B viral RNA in nasal and nasopharyngeal swabs from patients with signs and symptoms of respiratory infection. This test is intended for use as an aid in the differential diagnosis of influenza A and influenza B viral infections in humans in conjunction with clinical and epidemiological risk factors. The assay does not detect the presence of influenza C virus.

Negative results do not preclude influenza virus infection and should not be used as the sole basis for diagnosis, treatment or other patient management decisions.

Performance characteristics for influenza A were established during the 2011 and 2013 influenza seasons when influenza A/H3 and 2009 H1N1 influenza were the predominant influenza A viruses in circulation. When other influenza A viruses are emerging, performance characteristics may vary.

If infection with a novel influenza A virus is suspected based on current clinical and epidemiological screening criteria recommended by public health authorities, specimens should be collected with appropriate infection control precautions for novel virulent Influenza viruses and sent to state or local health department for testing. Viral culture should not be attempted in these cases unless a BSL 3+ facility is available to receive and culture specimens.

The assay can be performed using either the Life Technologies QuantStudio™ Dx, the Applied Biosystems® 7500 Fast Dx, or the Cepheid SmartCycler® II.

The Quidel Molecular Influenza A+B Assay detects viral nucleic acids that have been extracted from a patient sample using the NucliSENS® easyMAG® automated extraction platform. A multiplex real-time RT-PCR reaction is carried out under optimized conditions in a single tube generating amplicons for each of the target viruses present in the sample. This reaction is performed utilizing the Life Technologies QuantStudio™ Dx, the Applied Biosystems® 7500 Fast Dx, or the Cepheid SmartCycler® II platform. Identification of influenza A occurs by the use of target specific primers and a fluorescentlabeled probe that hybridizes to a conserved influenza A sequence within the matrix protein gene. Identification of influenza B occurs by the use of target specific primers and fluorescent-fabeled probes that will hybridize to a conserved influenza B sequence within the neuraminidase gene.

The Quidel Molecular Influenza A + B Assay is a multiplex Real Time RT-PCR assay for the in vitro qualitative detection and differentiation of influenza A and influenza B viral RNA in nasal and nasopharyngeal swabs from patients with signs and symptoms of respiratory infection. This device is intended to be used as an aid in the differential diagnosis of influenza A and influenza B viral infections in humans in conjunction with clinical and epidemiological risk factors.

Here's an analysis of the acceptance criteria and study as presented in the document:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implied by the clinical performance study aiming to demonstrate substantial equivalence to an FDA-cleared RT-PCR influenza detection device. The agreement metrics (Positive Percent Agreement, Negative Percent Agreement) are the key performance indicators. The document does not explicitly state pre-defined acceptance thresholds (e.g., "PPA must be > 90%"). However, the reported performance is presented in comparison to a predicate device.

2. Sample Size and Data Provenance for the Test Set

• Sample Size:
  • Initial specimens: 631 fresh swab specimens
  • Specimens remaining after exclusions for invalid results: 619
• Data Provenance: Prospective study conducted during the 2013 respiratory virus season (January to March 2013) at three sites across the United States.

3. Number of Experts and Qualifications for Ground Truth

The document does not explicitly state that experts were used to establish the ground truth for the clinical test set. Instead, a "high performance FDA-cleared Influenza A and B molecular test" was used as the comparator (reference standard) to determine agreement.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method for reconciling discordant results between the subject device and the comparator device. It simply reports the raw agreement percentages and notes the number of discordant dual infections.

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

No MRMC comparative effectiveness study was done. This study focuses on a standalone diagnostic device performance.

6. Standalone Performance

Yes, a standalone performance study was conducted. The "Clinical Performance" section evaluates the Quidel Molecular Influenza A + B Assay's performance (algorithm only, as it's an in vitro diagnostic test) against an FDA-cleared comparator device without human intervention in the result interpretation from the device itself.

7. Type of Ground Truth Used

The ground truth for the clinical test set was established by a "high performance FDA-cleared Influenza A and B molecular test" (a comparator device). This is a type of reference standard comparison, where the performance of the new device is measured against an established, cleared diagnostic method.

8. Sample Size for the Training Set

The document does not specify a separate training set or its sample size for the clinical performance evaluation. The clinical study described appears to be a validation/test set. For analytical performance (e.g., Limit of Detection, Analytical reactivity), quantified cultures of various influenza strains were used, but these are for analytical validation, not for training a machine learning model. This is a molecular diagnostic assay, not an AI/ML-based device in the sense of requiring a "training set" of patient data for model development.

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

As noted above, the concept of a "training set" for an AI/ML model with associated ground truth from patient data is not applicable here because this is a molecular diagnostic assay. For the analytical studies (e.g., Limit of Detection, Inclusivity), the "ground truth" was established by using quantified cultures of known influenza A and B strains at specified concentrations (TCID50/mL). These cultures serve as the known positive and negative controls at defined viral loads.

Ask a specific question about this device

The Quidel® Molecular Influenza A+B assay is a multiplex Real Time RT-PCR assay for the in vitro qualitative detection and differentiation of influenza A and influenza B viral RNA in nasal and nasopharyngeal swabs from patients with signs and symptoms of respiratory infection. This test is intended for use as an aid in the differential diagnosis of influenza A and influenza B viral infections in humans in conjunction with clinical and epidemiological risk factors. The assay does not detect the presence of influenza C virus.

Negative results do not preclude influenza virus infection and should not be used as the sole basis for diagnosis, treatment or other patient management decisions.

Performance characteristics for influenza A were established during the 2010 to 2011 influenza season when influenza A/H3 and 2009 H1N1 influenza were the predominant influenza A viruses in circulation. When other influenza A viruses are emerging, performance characteristics may vary.

If infection with a novel influenza A virus is suspected based on current clinical and epidemiological screening criteria recommended by public health authorities, specimens should be collected with appropriate infection control precautions for novel virulent Influenza viruses and sent to state or local health department for testing. Viral culture should not be attempted in these cases unless a BSL 3+ facility is available to receive and culture specimens.

The Quidel Molecular Influenza A+B Assay detects viral RNA that have been extracted from a patient sample using the NucliSENS® easyMAG® automated extraction platform. A multiplex RT-PCR reaction is carried out under optimized conditions in a single tube generating amplicons for each of the target viruses present in the sample. This reaction is performed utilizing the Cepheid SmartCycler® II platform. Identification of influenza A occurs by the use of target specific primers and a fluorescent- labeled probe that hybridizes to a conserved influenza A sequence within the matrix protein gene. Identification of influenza B occurs by the use of target specific primers and fluorescentlabeled probes that will hybridize to a conserved influenza B sequence within the neuraminidase gene.

The following is a summary of the procedure:

1. Sample Collection: Obtain nasal swabs and nasopharyngeal swabs specimens using standard techniques from symptomatic patients. These specimens are transported, stored, and processed according to established laboratory procedures.
2. Nucleic Acid Extraction: Extract Nucleic Acids from the specimens with the NucliSENS easyMAG System following the manufacturer's instructions using the appropriate reagents.

Prior to the extraction procedure add 20 µL of the Process Control (PRC) to each 180 uL aliquot of specimen. The PRC serves to monitor inhibitors in the extracted specimen, assures that adequate amplification has taken place and that nucleic acid extraction was sufficient.

3. Rehydration of Master Mix: Rehydrate the lyophilized Master Mix using 135uL of Rehydration Solution. The Master Mix contains oligonucleotide primers. fluorophore and quencher-labeled probes targeting highly conserved regions of the influenza A and influenza B viruses as well as the process control sequence. The primers are complementary to highly specific and conserved regions in the genome of these viruses. The probes are dual labeled with a reporter dye attached to the 5'end and a quencher attached to the 3' end. The rehydrated Master Mix is sufficient for eight reactions.
4. Nucleic Acid Amplification and Detection: Add 15 µL of the rehydrated Master Mix to each reaction tube. SuL of extracted nucleic acids (specimen with PRC) is then added to the tube. Then place the tube into the Cepheid SmartCycler® II.

Once the reaction tubes are added to the instrument, the assay protocol is initiated. This protocol initiates reverse transcription of the RNA targets generating complementary DNA, and the subsequent amplification of the target amplicons occurs. The Quidel Molecular Influenza A+B assay is based on TaqMan® chemistry, and uses an enzyme with reverse transcriptase, DNA polymerase, and 5'-3' exonuclease activities. During DNA amplification, this enzyme cleaves the probe bound to the complementary DNA sequence, separating the quencher dye from the reporter dye. This step generates an increase in fluorescent signal upon excitation by a light source of the appropriate wavelength. With each cycle, additional dye molecules are separated from their quenchers resulting in additional signal. If sufficient fluorescence is achieved by 45 cycles, the sample is reported as positive for the detected nucleic acid.

Here's a breakdown of the acceptance criteria and the study results for the Quidel Molecular Influenza A+B Assay, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the results of the comparative clinical study against an FDA-cleared predicate device. The performance characteristics are presented as Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA).

Note: The document explicitly states "good positive and negative percent agreement when compared to a high performance FDA Cleared Influenza A and B molecular test" in the conclusions, which serves as the general acceptance criterion. The precise numerical thresholds for "good" are not explicitly defined but are demonstrated by the presented results.

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

• Prospective Clinical Study:
  • Sample Size: 779 fresh specimens (427 nasal swabs and 352 nasopharyngeal swabs). After removing invalid specimens, 753 specimens were analyzed.
  • Data Provenance: The study was conducted during the 2010-2011 respiratory virus season (January to March 2011) at thirteen sites across the United States. The specimens were collected for routine influenza testing and tested at one central location within 72 hours of collection. This is prospective data.
• Retrospective Clinical Study:
  • Sample Size: 356 frozen nasopharyngeal swab specimens. After removing invalid specimens, 352 specimens were analyzed.
  • Data Provenance: The study used frozen specimens collected during the 2010-2011 respiratory virus season (January to March of 2011). This is retrospective data. The country of origin is not explicitly stated but implied to be the United States, similar to the prospective study, as it's part of the same submission to the FDA in the US.

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 studies was established using a "high performance FDA Cleared Influenza A and B molecular test" as the comparator method. Thus, it was not established by human experts in the typical sense (e.g., radiologist consensus), but rather by the performance of an already-cleared diagnostic device.

For the discordant results in the prospective study, sequence analysis was used to resolve discrepancies for influenza A (8 specimens that were negative by comparator but positive by Quidel Molecular) and influenza B (26 specimens negative by comparator but positive by Quidel Molecular, and 2 specimens negative by comparator and negative by sequence analysis). The qualifications of those performing the sequence analysis are not detailed.

4. Adjudication Method for the Test Set

For the clinical studies, results were compared directly against the predicate FDA-cleared RT-PCR device. In cases of discordance between the subject device and the comparator device, sequence analysis was performed for some discrepant specimens (specifically, those where the comparator was negative but the Quidel Molecular assay was positive for Influenza A or B). This indicates a form of adjudication where a third, more definitive method (sequencing) was used to assess the comparator's accuracy in those specific cases, rather than an expert panel reviewing the results.

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, an MRMC comparative effectiveness study involving human readers and AI assistance was not done. This submission is for a molecular diagnostic assay (RT-PCR), not an AI-powered image analysis or diagnostic tool that would typically involve human "readers." The comparison is between two molecular tests.

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

Yes, the device was evaluated in a standalone manner. The clinical performance data (Tables 5.8, 5.9, 5.10, 5.11) represent the Quidel Molecular Influenza A+B Assay's performance (algorithm only) compared to another FDA-cleared molecular test.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

The primary "ground truth" for the clinical performance evaluation was the results of a "high performance FDA Cleared Influenza A and B molecular test" (the predicate comparator device). For specific discordant results, sequence analysis was used as a more definitive method to further evaluate the initial test results.

8. The Sample Size for the Training Set

The document does not explicitly state a sample size for a "training set" in the context of machine learning or AI. This is a molecular diagnostic assay that functions through specific primer and probe binding, not a learning algorithm that requires a distinct training and test set in the AI sense.

However, the analytical performance studies (Limit of Detection, Analytical Reactivity, Analytical Specificity) involve testing various strains and concentrations, which could be considered akin to "training" or "development" data in a broader sense for optimizing assay parameters. For example:

• LoD study: Replicates of 20 per concentration for 3 influenza A strains and 3 influenza B strains.
• Analytical Reactivity: 38 influenza A strains and 15 influenza B strains, tested in triplicate.
• Analytical Specificity: Panels of 26 viral, 24 bacterial, and 1 yeast strain, tested in triplicate.

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

As this is not an AI/ML device with a conventional "training set," the concept of "ground truth for the training set" as it applies to AI is not directly applicable.

Instead, for the analytical studies:

• Limit of Detection (LoD): Ground truth was established by using quantified (TCID50/mL) cultures of known influenza strains serially diluted in negative matrix. The known concentration was the "truth."
• Analytical Reactivity (Inclusivity): Ground truth was established by using known, well-characterized viral strains (Influenza A subtypes and Influenza B strains) at specified TCID50 levels. The knowledge of the specific viral strain and its presence was the "truth."
• Analytical Specificity (Cross-reactivity): Ground truth was established by using known concentrations of specific viral, bacterial, and yeast strains. The knowledge of which organism was present (or not present, if testing for Influenza A/B) was the "truth."

Ask a specific question about this device

The Quidel Molecular Influenza A+B assay is a multiplex Real Time RT-PCR assay for the in vitro qualitative detection and differentiation of influenza A and influenza B viral RNA in nasal and nasopharyngeal swabs from patients with signs and symptoms of respiratory infection. This test is intended for use as an aid in the differential diagnosis of influenza A and influenza B viral infections in humans in conjunction with clinical and epidemiological risk factors. The assay does not detect the presence of influenza C virus.

Negative results do not preclude Influenza virus infection and should not be used as the sole basis for diagnosis, treatment or other patient management decisions.

Performance characteristics for influenza A were established during the 2010-2011 influenza season when influenza A/H3 and 2009 H1N1 influenza were the predominant influenza A viruses in circulation. When other influenza A viruses are emerging, performance characteristics may vary.

If infection with a novel influenza A virus is suspected based on current clinical and epidemiological screening criteria recommended by public health authorities, specimens should be collected with appropriate infection control precautions for novel virulent Influenza viruses and sent to state or local health department for testing. Viral culture should not be attempted in these cases unless a BSL 3+ facility is available to receive and culture specimens.

The Quidel Molecular Influenza A+B Assay detects viral nucleic acids that have been extracted from a patient sample using the NucliSENS® easyMAG® automated extraction platform. A multiplex real-time RT-PCR reaction is carried out under optimized conditions in a single tube generating amplicons for each of the target viruses present in the sample. This reaction is performed utilizing the Applied Biosystems® 7500 Fast Dx platform. Identification of influenza A occurs by the use of target specific primers and a fluorescentlabeled probe that hybridizes to a conserved influenza A sequence within the matrix protein gene. Identification of influenza B occurs by the use of target specific primers and fluorescent-labeled probes that will hybridize to a conserved influenza B sequence within the neuraminidase gene.

The following is a summary of the procedure:

1. Sample Collection: Obtain nasal swab and nasopharyngeal swab specimens using standard techniques from symptomatic patients. These specimens are transported, stored, and processed according to established laboratory procedures.
2. Nucleic Acid Extraction: Extract Nucleic Acids from the specimens with the NucliSENS easyMAG System following the manufacturer's instructions using the appropriate reagents. Use of other extraction systems with the Quidel Molecular Influenza A+B kit has not been validated. Validation of these systems is the responsibility of the end-user. Prior to the extraction procedure add 20 uL of the Process Control (PRC) to each 180 uL aliquot of specimen. The PRC serves to monitor inhibitors in the extracted specimen, assures that adequate amplification has taken place and that nucleic acid extraction was sufficient.
3. Rehydration of Master Mix: Rehydrate the lyophilized Master Mix using 135uL of Rehydration Solution. The Master Mix contains oligonucleotide primers, fluorophore and quencher-labeled probes targeting highly conserved regions of the influenza A and influenza B viruses as well as the process control sequence. The primers are complementary to highly specific and conserved regions in the genome of these viruses. The probes are dual labeled with a reporter dye attached to the 5-end and a quencher attached to the 3'-end. The rehydrated Master Mix is sufficient for eight reactions.
4. Nucleic Acid Amplification and Detection: Add 15 uL of the rehydrated Master Mix to each reaction plate well. 5uL of extracted nucleic acids (specimen with PRC) is then added to the plate well. Then place the plate into the ABI 7500 FastDx.

Once the plate is added to the instrument, the assay protocol is initiated. This protocol initiates reverse transcription of the RNA targets generating complementary DNA, and the subsequent amplification of the target amplicons occur. The Quidel Molecular Influenza A+B assay is based on TaqMan® chemistry, and uses an enzyme with reverse transcriptase, DNA polymerase, and 5'-3' exonuclease activities. During DNA amplification, this enzyme cleaves the probe bound to the complementary DNA sequence, separating the quencher dye from the reporter dye. This step generates an increase in fluorescent signal upon excitation by a light source of the appropriate wavelength. With each cycle, additional dye molecules are separated from their quenchers resulting in additional signal. If sufficient fluorescence is achieved by 35 cycles during the data collection stage of amplification, the sample is reported as positive for the detected nucleic acid.

This document describes the Quidel Molecular Influenza A+B Assay, a real-time RT-PCR assay for the qualitative detection and differentiation of influenza A and influenza B viral RNA.

Here's an analysis of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly stated as numerical targets in a single table, but rather are implied by the results presented in the analytical and clinical performance sections when compared to a legally marketed predicate device (Gen-Probe Prodesse ProFlu+).

Here’s a table summarizing the reported device performance, which implicitly suggests the acceptance criteria were met by demonstrating adequate performance comparable to the predicate regarding various analytical and clinical characteristics.

• High Negative (1.44E+01 TCID50/mL): 12/90 positive results across 3 sites, with AVE Ct ~34 and %CV ~2.0.
• Low Positive (9.6E+01 TCID50/mL): 90/90 positive across 3 sites, with AVE Ct ~27-29 and %CV ~3.5-7.0.
• Med Positive (2.4E+02 TCID50/mL): 90/90 positive across 3 sites, with AVE Ct ~25-27 and %CV ~2.9-5.5.
• Negative: 0/90 positive.
  Influenza B:
• High Negative (1.3E+01 TCID50/mL): 3/90 positive results across 3 sites, with AVE Ct 34.2 and %CV 1.2.
• Low Positive (8.6E+01 TCID50/mL): 90/90 positive across 3 sites, with AVE Ct ~24-25 and %CV ~2.6-5.1.
• Med Positive (2.2E+02 TCID50/mL): 90/90 positive across 3 sites, with AVE Ct ~22-23 and %CV ~2.0-2.9.
• Negative: 0/90 positive.
  Positive Controls for both A & B were 90/90 positive with low %CV (1.1-3.1). Conclusion: "generates reproducible results". |
  | Limit of Detection (LoD) | 95% of replicates test positive at the lowest concentration. | LoD for Influenza A strains ranged from 1.60E+01 to 9.20E+01 TCID50/mL.
  LoD for Influenza B strains ranged from 5.70E+00 to 4.30E+01 TCID50/mL.
  These values are within a similar range or better than the predicate device's LoD of 10² to 10⁻¹ TCID₅₀/mL. |
  | Analytical Reactivity (Inclusivity) | Detection of various influenza A and B strains at specified concentrations. | Detected 100% (38/38) of influenza A strains (including H1N1, 2009H1N1, H3N2, H5N1) and 100% (15/15) of influenza B strains at 10² to 10³ TCID50 levels. This included novel, pandemic, and avian influenza A strains and recent circulating influenza B strains. |
  | Analytical Specificity (Cross-reactivity) | No false positives with common respiratory pathogens or flora. | 100% analytical specificity. No cross-reactivity observed with 26 viral, 24 bacterial, and 1 yeast strain (all tested negative for Influenza A and B). |
  | Clinical Performance | High positive and negative percent agreement compared to a 510(k) cleared molecular device. | Prospective Study (N=668 fresh specimens):
• Influenza A: PPA 100% (139/139), NPA 98.5% (521/529)
• Influenza B: PPA 95.5% (105/110), NPA 97.8% (546/558)
  Retrospective Study (N=372 frozen specimens):
• Influenza A: PPA 100% (37/37), NPA 100% (335/335)
• Influenza B: PPA 97.4% (37/38), NPA 99.4% (332/334)
  Conclusion: "yielded good positive and negative percent agreement". |

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

• Test Set Sample Size:
  • Analytical Reproducibility: 90 replicates per level for each virus tested across 3 sites (30 per site).
  • Limit of Detection: Replicates of 20 per concentration of virus for each strain.
  • Analytical Reactivity: Triplicate testing for each of the 38 Influenza A strains and 15 Influenza B strains.
  • Analytical Specificity: Triplicate testing for each of the 26 viral, 24 bacterial, and 1 yeast strain.
  • Clinical Performance (Prospective): 668 fresh clinical specimens (373 nasal swabs, 313 nasopharyngeal swabs) after removing invalid results.
  • Clinical Performance (Retrospective): 372 frozen nasopharyngeal swabs after removing invalid results.
• Data Provenance: The document does not explicitly state the country of origin for the clinical samples. The study involved three laboratory sites for reproducibility, which might imply multi-site data collection, potentially from different locations. The clinical studies (prospective and retrospective) are referenced as "clinical studies," implying patient samples. The analytical studies use cultured viral strains and negative matrix.

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

The concept of "experts" in this context refers to the methods and outcomes used to establish the true presence or absence of influenza in clinical samples.

• Clinical Studies Ground Truth: The ground truth for the clinical test set was established by comparing the Quidel Molecular Influenza A+B Assay to a "Comparator: FDA Cleared RT-PCR device" (specifically identified as the Gen-Probe Prodesse ProFlu+ in the device comparison). For discordant results in the prospective study, sequence analysis was used to resolve discrepancies for Influenza A (7 cases) and Influenza B (12 cases). This suggests that a more definitive molecular testing method was considered the "expert" or "gold standard" for resolving these cases, rather than a human expert diagnosis.
• No explicit mention of human experts' qualifications for establishing ground truth, as the ground truth was primarily based on a predicate molecular test and confirmatory sequencing.

4. Adjudication Method for the Test Set

• For the clinical performance studies, the primary comparison was between the Quidel Molecular Influenza A+B Assay and the FDA Cleared RT-PCR device (predicate).
• Discordant results in the prospective clinical study were subjected to sequence analysis for adjudication:
  • For Influenza A: 7 specimens negative by the predicate but positive by the subject device were confirmed positive by sequence analysis. 1 specimen negative by both was also negative by sequence analysis.
  • For Influenza B: 12 specimens negative by the predicate but positive by the subject device were confirmed positive by sequence analysis.
  • This is a form of resolution by a higher-tier method rather than a traditional expert consensus method (e.g., 2+1 or 3+1 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

• No, an MRMC comparative effectiveness study was NOT done. This device is a molecular diagnostic assay, not an imaging or interpretive AI-based diagnostic tool that would involve human "readers" or "interpreters." The performance is evaluated based on the analytical and clinical accuracy of the assay itself.

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

• Yes, the studies evaluate the standalone performance of the assay. The Quidel Molecular Influenza A+B Assay is a laboratory-based real-time RT-PCR test. Its "performance" refers to the accuracy of the assay in detecting viral RNA in a sample. The results presented (reproducibility, LoD, inclusivity, specificity, clinical agreement) are all measures of the device's inherent performance. Human involvement is in sample collection, processing, and executing the assay protocol, but the "performance" itself is the output of the automated assay.

7. The Type of Ground Truth Used (Expert Consensus, Pathology, Outcomes Data, etc.)

• Clinical Ground Truth: Primarily established by comparison to a legally marketed, FDA-cleared molecular diagnostic device (Gen-Probe Prodesse ProFlu+). For discordant results, molecular sequence analysis was used as the confirmatory "gold standard" to establish true positivity/negativity.
• Analytical Ground Truth: For LoD, inclusivity, and specificity studies, the ground truth was based on quantified viral cultures (TCID50/mL) or bacterial/yeast cultures (CFU/mL), which are established analytical standards for concentration and identity.

8. The Sample Size for the Training Set

• The document does not specify a separate "training set" as would be typical for machine learning algorithms. This is a traditional molecular diagnostic assay where performance is characterized through analytical and clinical validation studies. The "development" of the assay involves optimization of primers, probes, and reaction conditions, but not typically a labeled training data set in the sense of AI.

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

• Since there isn't a "training set" explicitly mentioned or evaluated in the context of this traditional diagnostic device, the question of how its ground truth was established is not applicable. The assay's components and parameters would have been developed and optimized using well-characterized viral strains and clinical samples based on established laboratory methods.

Ask a specific question about this device