Search Results

The Acuitas® AMR Gene Panel, performed on the QIAGEN® EZ1® Advanced XL System and the OpGen Qualified QuantStudio™ 5 Real-Time PCR System, is a qualitative nucleic acid-based multiplex in vitro diagnostic test for detection and differentiation of antibiotic resistance markers to one or more antimicrobial agents. The test utilizes realtime polymerase chain reaction (PCR) and is performed on isolated colonies of Pseudomonas aeruginosa. Enteroooccus faecalis, or members of Enterobacterales grown in pure culture on blood agar or MacConkey agar.

Organism identification results must be available prior to reporting results for the Acuitas AMR Gene Panel. Antimicrobial resistance gene results are reported by the Acuitas AMR Gene Panel for the combinations of bacterial pathogens and associated genetic resistance markers indicated in Table 1 below.

The Acuitas AMR Gene Panel includes assays for the detection and reporting of genetic resistance markers associated with resistance to select drugs in the following antibiotic groups: aminoglycosides, carbapenems, cephalosporins, fluoroquinolones, penicillins, suffonamides, trimethoprim, and vancomycin, to aid in the identification of potentially antimicrobial-resistant organisms. The panel includes an assay for the detection of the mobilized colistin genetic determinant MCR-1, a marker of public health importance associated with reduced inhibitory activity of polymyxins.

The results of the Acuitas AMR Gene Panel for detection and identification of genetic determinants associated with antimicrobial resistance are used along with the Acuitas AMR Gene Panel Electronic User Guide (EUG). In certain cases, this information may be used as an aid to clinicians in the management of patients with known or suspected antibiotic non-susceptible or resistant bacterial infections. The EUG contains information on the appropriateness of reporting resistance markers detected by the Acuitas AMR Gene Panel for claimed organisms based on the strength of the collective, totality of scientific evidence delineating the level of association between molecular marker detection with phenotypic, clinical resistance. Test results are not conclusive or prescriptive for labeled use of any specific antimicrobial drug product, and therefore, this test cannot be used in place of or to postpone or delay phenotypic antimicrobial susceptibility testing.

A "Detected" or "Not Detected" result does not rule out the presence of other antimicrobial resistance markers not detected by the Acuitas AMR Gene Panel. A "Not Detected" result for a genetic marker of antimicrobial resistance does not indicate susceptibility to associated antimicrobial drugs or drug classes, as multiple mechanisms of resistance may exist.

The Acuitas® AMR Gene Panel is a qualitative nucleic acid-based in vitro diagnostic test capable of simultaneous detection and identification of select genetic determinants of antimicrobial resistance (AMR) in isolated bacterial colonies grown on blood agar or MacConkey agar. The test detects twenty-eight (28) genetic determinants of resistance to the following nine (9) antibiotic classes: aminoglycosides, carbapenems, cephalosporins, fluoroquinolones, penicillins, polymyxins, sulfonamides, trimethoprim and vancomycin. The assay is performed on pure colonies of Enterobacterales and Pseudomonas aeruginosa grown on blood agar or MacConkey agar along with Enterococcus faecalis grown on blood agar.

The Acuitas AMR Gene Panel kit contains all of the necessary reagents for PCR and detection in order to amplify and detect DNA from pure colonies of Enterobacterales (Citrobacter freundii complex (Citrobacter braakii, Citrobacter freundii, Citrobacter werkmanii, Citrobacter youngae), Citrobacter koseri, Enterobacter cloacae complex (Enterobacter asburiae, Enterobacter cloacae, Enterobacter hormaechei, Enterobacter kobei, Enterobacter ludwigiì), Escherichia coli, Klebsiella pneumoniae, Klebsiella quasipneumoniae, Klebsiella aerogenes, Klebsiella michiganensis, Klebsiella oxytoca, Klebsiella variicola, Morganii, Proteus mirabilis, Providencia rettgeri, Providencia stuartii, Raoultella ornithinolytica, Raoultella planticola, Serratia marcescens) and Pseudomonas aeruginosa grown on blood agar or MacConkey agar, as well as Enterococcus faecalis grown on blood agar from clinical specimens. The test kit includes PCR plates (96-well) with dried primers and probes for analysis of four (4) isolates (24 wells per isolate).

The Acuitas AMR Gene Panel assay employs automated deoxyribonucleic acid (DNA) extraction on the QIAGEN® EZ1® Advanced XL System and multiplex real-time PCR on an OpGen Qualified Applied Biosystems™ QuantStudio™ 5 Real-Time PCR System ("OpGen Qualified QuantStudio 5") for use with the Acuitas AMR Gene Panel. The QIAGEN EZ1 DSP Virus Kit has been selected for use as the sample preparation method for the Acuitas AMR Gene Panel test.

After colony isolation, the user prepares a 0.5 McFarland suspension for each bacterial isolate and performs DNA extraction. DNA is extracted on the QIAGEN EZ1 Advanced XL System according to manufacturer instructions incorporating the Assay Control within the extraction process. A sample of extracted DNA eluate is transferred to a Reagent Reservoir trough to which Acuitas AMR Gene Panel Master Mix is added. Extracted DNA/Master Mix is transferred to each of 24 wells on the Acuitas AMR Gene Panel PCR plate per test sample. The contents of each well are mixed, and the plate is sealed and transferred to the OpGen Qualified QuantStudio 5 for use with the Acuitas AMR Gene Panel for real-time multiplex reaction and detection using the Acuitas AMR Gene Panel PCR Template File.

Data are exported from the OpGen Qualified QuantStudio 5 and imported into the Acuitas AMR Gene Analysis Software, a spreadsheet application that analyzes the data and generates a report for viewing and printing. Each test report indicates detection of applicable antimicrobial resistance gene variants as "Detected", "Not Detected" or "NA/NR".

The Applied Biosystems QuantStudio 5 Real-Time PCR System is not intended for clinical diagnostic purposes. The OpGen Qualified QuantStudio 5 for use with the Acuitas AMR Gene Panel is a component of the Acuitas AMR Gene Panel assay and is cleared for in vitro diagnostic use only with the Acuitas AMR Gene Panel and not for any other application. The OpGen Qualified QuantStudio 5 for use with the Acuitas AMR Gene Panel may only be used with the Acuitas AMR Gene Panel after the instrument has been qualified for use by OpGen, Inc.

Here's a summary of the acceptance criteria and study details for the OpGen Acuitas AMR Gene Panel, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The document provides performance metrics primarily in terms of Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA) compared to Whole Genome Sequencing (WGS) for each AMR gene target. While explicit "acceptance criteria" values (e.g., "PPA must be >95%") are not explicitly stated as numbered criteria, the reported performance values implicitly serve as fulfilled acceptance criteria given the 510(k) clearance.

2. Sample Size for Test Set and Data Provenance

The "Clinical Performance Evaluation" used a total of 1,307 isolate samples.

• 1,224 clinical stock isolate samples: These were previously collected and de-identified. The document specifies that they were selected based on identification as Enterobacterales, Pseudomonas aeruginosa, or Enterococcus faecalis and documented non-susceptibility to at least one antibiotic class (excluding intrinsic resistance). The country of origin for these retrospective samples is not explicitly stated but implies a clinical setting (likely within the US given the FDA submission).
• 83 prospective isolate samples: These were unique and enrolled at clinical sites. These isolates were also selected based on their identification to the target organisms and documented non-susceptibility to at least one of the specified antibiotic classes. The country of origin is not explicitly stated but suggests a clinical setting.
• Total unique strains: 591 were evaluated (across all gene targets), with some strains tested in replicate, leading to the 1,307 total samples.

3. Number of Experts and Qualifications for Ground Truth

The document does not mention the use of human experts to establish ground truth in the traditional sense of medical image interpretation or clinical diagnosis. The ground truth for AMR gene detection was established using Whole Genome Sequencing (WGS), which is a highly objective, molecular-level method.

For organism identification, MALDI-TOF MS and automated species identification methods were used. These are standard laboratory techniques and do not involve human expert consensus for "ground truth" in the same way, but rather rely on established molecular and biochemical profiles.

4. Adjudication Method for Test Set

The document does not describe an explicit adjudication method (like 2+1 or 3+1) for the clinical performance evaluation, as the primary ground truth for AMR gene detection was WGS. WGS is considered a definitive molecular method, rendering such adjudication methods unnecessary for this type of test.

However, in the "Media Equivalency" study (Table 21 footnotes, particularly footnote b), it notes an "adjudication testing" event: "Adjudication testing was performed with this isolate in duplicate on both agar media and true positives for CTX-M-1, RMT and TEM were obtained for all samples." This suggests that for potentially discordant results in specific analytical studies, further internal verification or re-testing (which they call "adjudication testing") was performed, rather than external expert consensus.

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

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. This device is an in vitro diagnostic (IVD) test for detecting specific genetic markers, not an AI-assisted diagnostic tool for human readers. Therefore, there is no "human readers improve with AI vs without AI assistance" effect size to report.

6. Standalone (Algorithm Only) Performance

Yes, the primary performance evaluation presented is the standalone performance of the Acuitas AMR Gene Panel. The reported PPA and NPA values in "Clinical Performance" (Table 24) compare the device's output (detection of AMR genes) directly against Whole Genome Sequencing (WGS), which serves as the independent and highly accurate ground truth. This is a direct measure of the algorithm/device's performance without human intervention in the interpretation of the results.

7. Type of Ground Truth Used

The primary ground truth for AMR gene detection in the clinical study was Whole Genome Sequencing (WGS).
Other ground truth methods used in the broader studies (clinical and analytical) include:

• MALDI-TOF MS for organism identification (Gram-positive).
• Automated species identification methods for organism identification (Gram-negative).
• Antimicrobial Susceptibility Testing (AST) by broth microdilution to determine phenotypic susceptibility (used for context and selection criteria of isolates, but WGS was the direct comparator for gene detection).

8. Sample Size for Training Set

The document does not explicitly state a sample size for a "training set" in the context of machine learning or AI models with a distinct training phase. This device is a real-time PCR assay and its development would typically involve design and validation of primers/probes and optimization against a collection of known target and non-target sequences/isolates. The "Analytical Reactivity (Inclusivity)" study (298 isolates) and "Analytical Specificity (Cross-Reactivity)" study (423 isolates) describe testing against a comprehensive set of known gene variants and non-target organisms, which collectively contribute to the robustness of the assay design (analogous to a very targeted training/validation approach for molecular assays).

9. How Ground Truth for Training Set Was Established

Given that this is a molecular diagnostic assay (PCR-based) and not an AI/machine learning model in the typical sense, there isn't a "training set" ground truth established in the same manner. The assay is designed to detect specific gene sequences. The "ground truth" for developing such an assay typically relies on:

• Known gene sequences: Public databases (e.g., NCBI) for AMR genes.
• Sequencing (WGS/Sanger sequencing): To confirm the presence or absence of target genes in isolates used during assay development and optimization.
• Phenotypic AST: To correlate gene presence with resistance phenotypes, which helps in the clinical interpretation of the assay.

The analytical studies (inclusivity and specificity) served to validate the comprehensive detection capabilities against a wide range of molecular "ground truths" (confirmed by WGS) and non-targets.

Ask a specific question about this device

The GenePOC™ Carba assay, performed on the revogene™ instrument, is a qualitative in vitro diagnostic test designed for the detection and differentiation of the blakec. blaNDM, blackA-48-like, and blamp gene sequences associated with carbapenem-nonsusceptible pure colonies of Enterobacteriaceae, Acinetobacter baumannii, or Pseudomonas aeruginosa, when grown on blood agar or MacConkey agar. The test utilizes automated real-time Polymerase Chain Reaction (PCR).

The GenePOC™ Carba assay should be used in conjunction with other laboratory tests including phenotypic antimicrobial susceptibility testing. A negative GenePOC™ Carba assay result does not preclude the presence of other resistance mechanisms.

The GenePOC™ Carba assay is intended as an aid for infection control in the detection of carbapenem-non-susceptible bacteria that colonize patients in healthcare settings. The identification of a bland, blaym or blaviM metallo-B-lactamase gene (i.e., the genes that encode the IMP, NDM and VIM metallo-ß-lactamases, respectively) may be used as an aid to clinicians in determining appropriate therapeutic strategies for patients with known or suspected carbapenem non-susceptible infections.

The GenePOC™ Carba assay is an automated real-time polymerase chain reaction (PCR) in vitro diagnostic test for the qualitative detection and differentiation of the blaype, blayDM, blavin, blaoxA-48-ike, and blamp gene sequences in carbapenem-non-susceptible pure colonies of Enterobacteriaceae, Acinetobacter baumannii, or Pseudomonas aeruginosa grown on blood agar or MacConkey agar. The GenePOC™ Carba assay is designed to be performed on the revogene™ instrument. The revogene™ is an instrument that automates sample homogenization, sample dilution, cells lysis, DNA amplification, and detection of the amplified PCR products.

The GenePOC™ Carba assay kit is comprised of single-use, disposable microfluidic cartridges (PIEs), Sample Buffer Tubes (SBT), and Disposable Transfer Tools (DTT). These components are used to dilute the sample, extract, amplify, and simultaneously detect blayDM, blavin, blaoxA-48-like, and blamp DNA. A Process Control (PrC) is also incorporated into each PIE to verify sample processing and amplification steps. The PrC allows for the verification of potential inhibitor substances as well as microfluidic, instrument or reagent failure.

Each GenePOC™ Carba assay kit contains 24 individual pouches. Each pouch has components for one (1) test including one (1) Carba PIE, one (1) SBT, and one (1) DTT. The single-use, multi-chambered fluidic cartridges are designed to complete sample preparation and real-time PCR for the detection of the blaker, blands, blaoxa-48-like, and blanyp gene sequences from isolates of pure cultures of carbapenem-non-susceptible gram-negative bacteria in approximately 70 minutes. User intervention is only required for preparing the standardized 0.5 McFarland bacterial suspension from characterized carbapenem-non-susceptible isolated colonies, inoculating the bacterial suspension into the Sample Buffer Tube (SBT), transferring the sample into the single-use disposable PIE, and loading/unloading the PIEs into the revogene™ carousel. Each PIE is a completely integrated closed device in which a sample is dispensed and processed through different microfluidic chambers and channels that allow for the sample processing and subsequent real-time PCR steps.

Upon completion of a run, the results are interpolated by the revogene™ from measured fluorescent signals and embedded calculation algorithms. The output results include positive, negative, indeterminate, and unresolved. Upon completion of a run, the user removes the used cartridges and disposes of them in normal biological waste. The results are automatically generated at the end of the process in a report that can be viewed and printed.

Here's a breakdown of the acceptance criteria and the study details for the GenePOC™ Carba assay, based on the provided FDA 510(k) summary:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are implied by the clinical study's objective: to establish the performance characteristics (sensitivity and specificity) of the GenePOC™ Carba assay compared to a Reference Method for detecting carbapenemase genes. No explicit numerical acceptance criteria (e.g., "Sensitivity >= 95%") are formally stated in the provided text as pass/fail thresholds for clearance, but the reported performance represents what was achieved and found acceptable for substantial equivalence.

Table of Acceptance Criteria (Implied) and Reported Device Performance

The summary presents performance broken down by each target gene. The "Reference Method" in the clinical study refers to an FDA-cleared Nucleic Acid Amplification Test (NAAT) whose performance was established against PCR/bidirectional sequencing.

Performance on Isolated Colonies Grown on Blood Agar Relative to Reference Method (Overall by Target)

Performance on Isolated Colonies Grown on MacConkey Agar Relative to Reference Method (Overall by Target)

Study Details

1. Sample size used for the test set and the data provenance:

   • Test Set Size: 512 fully compliant bacterial isolates.
     • 475 clinical stock isolates.
     • 57 prospectively collected fresh isolates.
   • Data Provenance: Three clinical centers (one in Canada and two in the United States of America).

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

   • The ground truth was established by a "Reference Method" (an FDA-cleared Nucleic Acid Amplification Test) whose performance was previously established against PCR/bidirectional sequencing. The document does not specify the number or qualifications of experts involved in running or interpreting this reference method or the PCR/bidirectional sequencing, nor in the initial characterization of the clinical isolates.

3. Adjudication method for the test set:

   • Discordant analysis: For samples with discrepant results between the GenePOC™ Carba assay and the Reference Method, further testing was performed using alternative PCR for each of the five assay analytes, followed by bi-directional Sanger sequencing. The results of this discrepant testing were then used to reconcile the initial findings (effectively re-evaluating the "true" status of the sample).

4. 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 with/without AI assistance was not performed. This device is an automated in vitro diagnostic test, not an AI-assisted diagnostic imaging device targeting human interpretation.

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

   • Yes, the clinical study assessed the "standalone" performance of the GenePOC™ Carba assay. The device is an automated real-time PCR platform that extracts, amplifies, and detects target DNA, and "the results are interpolated by the revogene™ from measured fluorescent signals and embedded calculation algorithms." User interaction is primarily for sample preparation and loading/unloading. The reported sensitivity and specificity values represent this standalone algorithmic performance against the defined Reference Method.

6. The type of ground truth used:

   • The ground truth for the clinical study was established by a Reference Method, which was an FDA-cleared Nucleic Acid Amplification Test (NAAT). This NAAT itself had its performance previously established in comparison to PCR/bidirectional sequencing. For discordant results, alternative PCR followed by bi-directional Sanger sequencing was used for final adjudication, indicating that molecular sequencing data served as the ultimate ground truth for discrepant cases.

7. The sample size for the training set:

   • The document describes analytical and clinical performance studies, but it does not specify a separate "training set" for the algorithm's development in the context of machine learning. For an in vitro diagnostic like this, the "training" involves assay design, optimization, and establishment of thresholds. The document states, "The assay cut-offs were determined by testing a total of 599 samples" (Section N.1.g), which could be considered part of the internal development/training phase for establishing the algorithm's decision boundaries.

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

   • As noted above, for the "training" (assay cut-off determination), ground truth would have been established through well-characterized samples with known presence or absence of the target genes. The analytical reactivity (inclusivity) study also contributes to this, using 58 carbapenem-non-susceptible isolates with known resistance genes and variants, determined by their collection number, geographical/temporal origin, and reported resistance gene/variant. For the 599 samples used to determine assay cut-offs, the specific method for establishing their ground truth is not detailed, but it is implied to be based on the known genetic makeup of the strains, likely through sequencing or other established molecular methods.

Ask a specific question about this device

The Xpert® Carba-R Assay, performed on the GeneXpert® Instrument Systems, is a qualitative in vitro diagnostic test for the detection and differentiation of the blaKPC, blaNDM, blaVIM, blaOXA-48, and blaIMP gene sequences associated with carbapenem-non-susceptible pure colonies of Enterobacteriaceae, Acinetobacter baumannii, or Pseudomonas aeruginosa grown on blood agar or MacConkey agar. The test utilizes automated real-time polymerase chain reaction (PCR).

A negative Xpert Carba-R Assay result does not preclude the presence of other resistance mechanisms. The Xpert Carba-R Assay should be used in conjunction with other laboratory tests including phenotypic antimicrobial susceptibility testing. The Xpert Carba-R Assay is intended as an aid for infection control in detecting and differentiating genetic markers of resistance to monitor the spread of carbapenem-non-susceptible organisms in healthcare settings. The Xpert Carba-R Assay is not intended to guide or monitor treatment for carbapenem-non-susceptible bacterial infections.

The Xpert Carba-R Assay is an automated real-time polymerase chain reaction (PCR) in vitro diagnostic test for qualitative detection of the blaKPC, blaNDM, blaVIM, blaOXA-48, and blaIMP gene sequences from isolates of pure cultures of carbapenem-non-susceptibility gram-negative bacteria. The Xpert Carba-R Assay is intended as an aid for infection control for monitoring the spread of carbapenem-non-susceptible organisms in healthcare settings.

The Xpert Carba-R Assay is performed on the Cepheid GeneXpert® Instrument Systems (GeneXpert Dx, GeneXpert Infinity-48, GeneXpert Infinity-48s, and GeneXpert Infinity-80 systems). The GeneXpert Instrument System platform automates sample preparation, amplification and real-time detection.

The GeneXpert Instrument Systems require the use of single-use, disposable cartridges (the Xpert Carba-R cartridges) that hold the PCR reagents and host the PCR process. Because the cartridges are self-contained and specimens never come into contact with working parts of the instrument modules, cross-contamination between samples is minimized.

The Xpert Carba-R Assay cartridges contain reagents for the detection of blaKPC, blaNDM, blaVIM, blaOXA-48, and blaIMP gene sequences. A Sample Processing Control (SPC) and a Probe Check Control (PCC) are controls utilized by the GeneXpert Instrument System platform. The SPC is present to control for adequate processing of the target bacteria and to monitor the presence of inhibitors in the real-time PCR reaction to reduce the possibility of false negative results. The PCC verifies reagent rehydration, real-time PCR tube filling in the cartridge, probe integrity, and dye stability.

The single-use, multi-chambered fluidic cartridges are designed to complete sample preparation and real-time PCR for the detection of the blaKPC, blaNDM, blaVIM, blaOXA-48, and blaIMP gene sequences from isolates of pure cultures of carbapenem-nonsusceptibility gram-negative bacteria in approximately 50 minutes. The GeneXpert Instrument Systems, comprised of the GeneXpert Dx Systems and the GeneXpert Infinity Systems, have 1 to 80 randomly accessible modules, depending upon the instrument, that are each capable of performing separate sample processing and real-time PCR and RT-PCR tests. Each module contains a syringe drive for dispensing fluids (i.e., the syringe drive activates the plunger that works in concert with the rotary valve in the cartridge to move fluids between chambers), an ultrasonic horn for lysing cells or spores, and a proprietary I-CORE® thermocycler for performing real-time PCR and RT-PCR and detection.

The bacterial isolates from culture are placed into a sample reagent. The sample is transferred to the sample chamber of the disposable fluidic cartridge (the Xpert Carba-R cartridge). The user initiates a test from the system user interface and places the cartridge into the GeneXpert instrument platform, which performs hands-off real-time, multiplex PCR for detection of the blaKPC, blaNDM, blaVIM, blaOXA-48, and blaIMP gene sequences. The results are automatically generated at the end of the process in a report that can be viewed and printed.

Here's a summary of the acceptance criteria and study findings for the Xpert Carba-R Assay, based on the provided document:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state pre-defined acceptance criteria (e.g., "Sensitivity must be >90%"). Instead, it presents the results of the clinical study as evidence of the device's performance. For clarity, I will interpret the observed performance as the de facto acceptance criteria demonstrated by the study.

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

• Sample Size (Clinical Study Test Set): A total of 467 eligible isolates were included in the final dataset for clinical performance analysis. These consisted of 410 clinical stock isolates and 57 fresh isolates.
• Data Provenance: The isolates were sourced from a multi-site investigational study. The document does not specify the country of origin of the data, but the context of an FDA submission suggests it could be primarily US-based or from international sites participating in studies for FDA clearance. The study used both retrospective (clinical stock isolates) and prospective (fresh isolates) data.

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

The document does not explicitly state the "number of experts" or their specific "qualifications" for establishing ground truth. Instead, the ground truth for the clinical study was established by reference bi-directional sequencing of the amplified DNA target, which is a highly standardized and objective molecular method. For discordant results, discrepant testing was performed using the same bi-directional sequencing method. This means the "ground truth" relies on robust analytical methods rather than subjective expert consensus.

4. Adjudication Method for the Test Set

The document describes the following adjudication method:

• For isolates with discordant results between the Xpert Carba-R Assay and initial reference sequencing (performed locally), discrepant testing was performed using bi-directional sequencing on isolates from MacConkey agar plates, and the results are footnoted in the performance tables.
• For the initial reference sequencing, if "no bands were shown on the Bioanalyzer for any of the five target genes," the isolate was "not sent for sequence analysis and the reference method result was considered negative for the five target genes." If bands were shown, the amplicon was sent to an independent laboratory for reference bi-directional sequencing analysis.

This implies a hierarchical adjudication where the primary reference method is bi-directional sequencing, with a secondary review/confirmation using the same method for discordant 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

This question is not applicable to this device. The Xpert Carba-R Assay is an automated in vitro diagnostic test for detecting genetic sequences. It is a standalone algorithm (molecular assay) that provides a qualitative result (DETECTED/NOT DETECTED) and does not involve human readers interpreting images or data that an AI might assist with. Therefore, an MRMC comparative effectiveness study involving human readers and AI assistance was not performed.

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

Yes, a standalone study was done. The entire clinical performance evaluation described compares the Xpert Carba-R Assay's automated results directly against the reference standard (bi-directional sequencing). The device itself is designed for "hands-off real-time, multiplex PCR" with "results automatically generated." This inherently means its performance was evaluated as a standalone algorithm without human intervention in the result generation or interpretation phase beyond the initial sample preparation.

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

The primary type of ground truth used for the clinical performance study was molecular reference sequencing (bi-directional DNA sequencing) of the amplified DNA target. This is a highly objective and analytical method to determine the presence or absence of the target gene sequences.

8. The Sample Size for the Training Set

The document does not explicitly describe a separate "training set" for an algorithm in the way machine learning algorithms typically use them. This is a molecular diagnostic assay, not an AI/machine learning model. The initial development and optimization of the assay would have involved various analytical studies (e.g., analytical reactivity, analytical specificity, limit of detection, etc.) using numerous characterized strains, but these are not typically referred to as a "training set" in the context of molecular assays as they are for AI algorithms. The "Analytical Reactivity (Inclusivity)" study used a panel of 71 well-characterized bacterial strains.

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

As noted above, a traditional "training set" in the AI sense is not applicable here. For the analytical studies and assay development:

• Analytical Reactivity (Inclusivity): The ground truth for the 71 bacterial strains tested was established by their "well-characterized" nature and "variant information," likely through prior molecular characterization (e.g., sequencing) to confirm the presence and type of resistance markers.
• Analytical Specificity (Cross-reactivity): The ground truth for the 62 non-target bacterial strains and 24 commensal/enteric microorganisms was established by "Confirmed Resistance Mechanisms" (determined by individual PCR assays, DNA sequence analysis, or Check-Points array version CT102) or their known carbapenem susceptibility/resistance profiles.

These methods ensure the accurate molecular identification of the strains used in the analytical performance evaluations during the assay's development and validation.

Ask a specific question about this device