Search Results

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most Gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaceae and Non-Enterobacteriaceae and most Gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus, Enterococcus and Streptococcus.

This premarket notification is for the addition of the Phoenix Inducible Macrolide Resistance (iMLSb) Test in Staphylococcus species to Gram-positive ID/AST or AST only Phoenix panels. The Phoenix Inducible Macrolide Resistance Test is used to detect inducible macrolide-lincosamide-streptogramin B resistance in Staphylococcus species.

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. The system includes the following components:

• . BD Phoenix instrument and software.
• . BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents or AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum. .
• BD Phoenix AST Broth used for performing AST tests only. .
• BD Phoenix AST Indicator solution added to the AST Broth to aid in bacterial growth . determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrenc tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and preliminarily identified as a Gram-negative or Gram-positive isolate. Phoenix panels are inoculated with a specified organism density and placed into the instrument.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurcments of changes to the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35°C. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, R or N (susceptible, intermediate, resistant or not susceptible).

The provided text describes the BD Phoenix™ Automated Microbiology System for detecting inducible macrolide resistance (iMLSb) in Staphylococcus species. Here's a breakdown of the acceptance criteria and the study that proves the device meets them:

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

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

• Sample Size for Test Set:
  • Clinical Study Isolates: Not explicitly stated as a single number for the entire test set. However, for the Phoenix Inducible Macrolide Resistance Test, the reported "CA (n)" is 295, indicating 295 isolates were used for this specific test within the clinical study.
  • Site Reproducibility: A "panel of Gram-positive isolates" was used. Each site tested these isolates in triplicate on three different days. The exact number of isolates in this panel is not specified.
• Data Provenance: Clinical, stock, and challenge isolates were tested at multiple geographically diverse sites across the United States. The study type for clinical isolates was comparative against a reference method, which is typical for prospective data collection in such studies, though not explicitly stated. Challenge sets are typically retrospective (known results).

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

The document does not explicitly state the number of experts and their qualifications used to establish the ground truth for the test set. The ground truth for clinical isolates was established by comparing to the "CLSI reference broth microdilution method". For challenge isolates, results were compared to "expected results," implying a pre-defined ground truth. The interpretation of these reference methods would typically be performed by trained laboratory personnel, but no specific expert qualifications are provided.

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

The document does not describe any specific adjudication method (like 2+1 or 3+1) for resolving discrepancies in the test set. It mentions comparison to a reference method, suggesting that the reference method's result served as the definitive interpretation.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done in the context of human readers with or without AI assistance. This device is an automated system (BD Phoenix Automated Microbiology System) for in vitro susceptibility testing, not an AI-assisted diagnostic tool that aids human interpretation of images or other data. The comparison is between the automated system and a reference laboratory method.

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

Yes, this was a standalone performance study. The BD Phoenix Automated Microbiology System is an "algorithm only" type of device in the sense that it automatically performs the susceptibility testing and provides results (MIC values and category interpretations S, I, R, or N) without direct human interpretation within the measurement process itself. The study evaluates the performance of this automated system against reference methods.

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

The primary type of ground truth used was:

• Reference Method Results: For clinical isolates, the "CLSI reference broth microdilution method" was used as the ground truth. This is a laboratory-based, standardized method considered the gold standard for antimicrobial susceptibility testing.
• Expected Results: For challenge isolates, "expected results" were used. This typically refers to isolates with well-characterized resistance profiles, often obtained from reference laboratories or collections.

8. The sample size for the training set

The document does not specify a separate "training set" for the BD Phoenix Automated Microbiology System in the context of this 510(k) submission. This is an antimicrobial susceptibility testing (AST) system; its "training" fundamentally involves the development and calibration of the instrument and its reagent panels to accurately detect bacterial growth and interpret MICs according to established microbiological principles and CLSI guidelines. The performance data presented (site reproducibility and clinical studies) is primarily for verification/validation.

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

Since a distinct "training set" as understood in machine learning (where ground truth is typically assigned by experts for algorithm learning) is not explicitly mentioned or applicable in the traditional sense for this type of automated microbiology system, the method for establishing ground truth for a training set is not described. The system's underlying design and calibration would have relied on extensive microbiological knowledge and reference methods, ensuring its programmed interpretations align with CLSI standards.

Ask a specific question about this device

The Cepheid Xpert™ MRSA/SA Blood Culture Assay performed on the GeneXpert® Dx System™ is a qualitative in vitro diagnostic test intended for the detection of Staphylococcus aureus (SA) and methicillin-resistant Staphylococcus aureus (MRSA) DNA directly from patient positive blood cultures. The assay utilizes automated realtime polymerase chain reaction (PCR) for the amplification of MRSA/SA specific DNA targets and fluorogenic target-specific hybridization probes for the real-time detection of the amplified DNA. The assay is performed directly on positive blood culture specimens using BD BACTEC™ Plus Aerobic/F blood culture bottles that are determined as Gram Positive Cocci in Clusters (GPCC) or as Gram Positive Cocci in singles (GPC) by Gram stain. The Cepheid Xpert™ MRSA/SA Blood Culture Assay is not intended to monitor treatment for MRSA/SA infections. Subculturing of positive blood cultures is necessary to recover organisms for susceptibility testing or for epidemiological typing.

The Cepheid Xpert™ MRSA/SA Blood Culture Assay (Xpert MRSA/SA Blood Culture Assay) is a rapid, automated DNA test for simultaneously detecting MRSA and SA Alber) I a rapids, and culture specimens. The specimens. The specimen consists of an aliquot taken from a positive blood culture bottle for testing with the Xpert MRSA/SA Blood Culture Assay. Using one of the small disposable transfer pipettes provided with the test kit, a single drop of the positive blood culture aliquot (approximately 50 µL) is transferred into the Elution Reagent. The Elution Reagent is vortexed for 10 seconds and the entire contents are transferred to "S" chamber of the disposable fluidic cartridge (the Xpert MRSA/SA Blood Culture Assay cartridge). The two single-use reagents (Reagent 1 and Reagent 2) that are provided with the assay are transferred to different, uniquelylabeled chambers of the Xpert MRSA/SA cartridge. The user initiates a test from the system user interface and places the cartridge into the GeneXpert® Dx System instrument platform, which performs hands-off real-time, multiplex polymerase chain reaction (PCR) for detection of DNA. In this platform, additional sample preparation, amplification, and real-time detection are all fully-automated and completely integrated.

The GeneXpert Dx System consists of a GeneXpert instrument, personal computer, and the multi-chambered fluidic cartridges that are designed to complete sample preparation and real-time PCR for detection of MRSA and SA in approximately 50 minutes. Each system has 2 to 16 randomly accessible modules that are each capable of performing separate sample preparation and real-time PCR tests. Each module contains a syringe drive for dispensing fluids, an ultrasonic horn for lysing cells or spores, and a proprietary I-CORE® thermocycler for performing real-time PCR and detection.

The Xpert MRSA/SA Assay includes reagents for the simultaneous detection of the target organisms, SA and MRSA. The primers and probes in the Xpert MRSA/SA Assay detect nucleic acid sequences of the staphylococcal protein A (spa), the gene for methicillin/oxacillin resistance (mecA), and staphylococcal cassette chromosome (SCCmec) inserted in the SA chromosomal attB site.

The test includes a Sample Processing Control (SPC) to control for adequate processing of the target bacteria and to monitor the presence of inhibitor(s) in the PCR assay. The Probe Check Control (PCC) verifies reagent rehydration, PCR tube filling in the cartridge, probe integrity, and dye stability.

The acceptance criteria and study proving the device meets them are detailed below for the Cepheid Xpert™ MRSA/SA Blood Culture Assay.

1. Table of Acceptance Criteria and Reported Device Performance:

The document implicitly defines acceptance criteria based on the performance observed in the predicate device and the clinical study results. The clinical study results serve as the primary demonstration of meeting these criteria.

Note: The predicate device's performance is listed as "Positive % Agreement" and "Negative % Agreement," which are equivalent to sensitivity and specificity in this context. The implicit acceptance criteria for the new device are to perform at least comparably to the predicate and the "standard of care" (reference culture results and susceptibility testing).

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

• Sample Size for Clinical Test Set: 249 specimens.
• Data Provenance: Multi-site prospective investigation study at three US institutions. The data is prospective, collected from subjects whose routine care included blood culture testing.

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

The document does not explicitly state the number of experts or their specific qualifications (e.g., years of experience as a radiologist) used to establish the ground truth for the clinical test set. It mentions "reference culture results and susceptibility testing, the current standard of care" and that susceptibility testing was performed "in accordance with the CLSI documents M2-A9 and M100-S17." This implies that qualified laboratory personnel, following established clinical microbiology guidelines, performed these reference methods.

4. Adjudication Method for the Test Set:

The document does not explicitly describe an adjudication method (e.g., 2+1, 3+1). The ground truth was established by "reference culture results and susceptibility testing, the current standard of care." This suggests a direct comparison against the established laboratory gold standard, rather than a consensus among multiple human readers.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and Effect Size:

No, a multi-reader multi-case (MRMC) comparative effectiveness study was not explicitly done in the context of comparing human readers with and without AI assistance. The study compares the device's performance directly against "reference culture results and susceptibility testing," which is the established standard of care for identifying bacterial infections. The device is an automated nucleic acid amplification test, not an AI assistance tool for human interpretation.

6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done:

Yes, a standalone performance study was conducted. The Cepheid Xpert™ MRSA/SA Blood Culture Assay is an automated device that performs sample preparation, amplification, and real-time detection without human intervention once the sample is loaded. The "Overall Results" section (Page 12) directly reports the device's performance in identifying MRSA and SA specimens relative to culture. This refers to the algorithm's performance without a human in the loop for interpretation.

7. The Type of Ground Truth Used:

The ground truth used for the clinical study was based on:

• Reference Culture Results: This involves standard microbiological culture techniques to isolate and identify organisms.
• Susceptibility Testing: Performed in accordance with CLSI documents (M2-A9 and M100-S17) to determine methicillin/oxacillin resistance, using cefoxitin as a surrogate.

This represents established microbiological laboratory standards, often considered the "gold standard" for pathogen identification and antimicrobial susceptibility.

8. The Sample Size for the Training Set:

The document does not explicitly state a separate "training set" for the clinical validation studies in the way one might for a machine learning algorithm. For analytical studies, various strains were tested:

• Analytical Inclusivity Study on CDC Staphylococcus aureus Specimens: 25 Staphylococcus aureus strains.
• Analytical Inclusivity Study on Expanded Panel of Staphylococcus aureus Specimens: 121 additional Staphylococcus aureus strains.
• "Empty Cassette Variants" Study: 22 Staphylococcus aureus isolates.
• Analytical Limit of Detection: Replicates of 20 for 6 MRSA isolates and 3 MSSA isolates.
• Cross-reactivity Study: 105 strains (98 ATCC, 7 NARSA).

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

For the analytical "training" (or rather, validation) sets mentioned above, the ground truth was established through:

• Bacterial strain identification, PFGE type, and SCCmec type determined by the CDC (for CDC specimens).
• Catalase, tube coagulase, and Gram stain for characterizing discordants in the expanded panel study.
• Methicillin susceptibility assessed by disk diffusion using a cefoxitin disk.
• Known characteristics and genetic information of the cultured strains (e.g., MRSA, SA, Cfu/test).

Ask a specific question about this device

The Cepheid Xpert MRSA/SA Skin and Soft Tissues Infection Assay (Xpert MRSA/SA SSTI Assay) performed in the GeneXpert® Dx System is a qualitative in vitro diagnostic test intended for the detection of Staphylococcus aureus (SA) and methicillin-resistant Staphylococcus aureus (MRSA) from skin and soft tissue infection swabs. The test utilizes automated real-time polymerase chain reaction (PCR) to detect MRSA/SA DNA. The Xpert MRSA/SA SSTI Assay is indicated for use in conjunction with other laboratory tests such as microbiology culture, and clinical data available to the clinician as an aid in the detection of MRSA/SA from skin and soft tissue infections. The Xpert MRSA/SA SSTI Assay is not intended to monitor treatment for MRSA/SA infections. Concomitant cultures for SA and MRSA are necessary to recover organisms for susceptibility testing or epidemiological typing. In a mixed culture containing MRSA/SA and other organisms (e.g. Gram negative bacilli, yeast), results can be false negative or variable depending on the concentration of MRSA/SA present, particularly if the concentration of MRSA/SA is close to the LoD of the assay.

The Cepheid Xpert MRSA/SA Skin and Soft Tissue Infection Assay (Xpert MRSA/SA SSTI Assay) is a rapid, automated DNA test for simultaneously detecting MRSA and SA directly from skin and soft tissue specimen is collected on a double swab, which is placed in a tube containing elution reagent. Following brief vortexing, the eluted matcrial and two single-use reagents (Reagent 1 and Reagent 2) that are provided with the assay are transferred to different, uniquely-labeled chambers of the disposable fluidic cartridge (the Xpert MRSA/SA cartridge). The user initiates a test from the system user interface and places the cartridge into the GeneXpert® Dx System instrument platform, which performs hands-off real-time, multiplex polymerase chain reaction (PCR) for detection of DNA. In this platform, additional sample preparation, amplification, and real-time detection are all fully-automated and completely integrated. The GeneXpert® System consists of a GeneXpert instrument, personal computer, and the multi-chambered fluidic cartridges that are designed to complete sample preparation and real-time PCR for detection of MRSA and SA in approximately 50 minutes. Each system has 1 to 16 randomly accessible modules that are each capable of performing separate sample preparation and real-time PCR tests. Each module contains a syringe drive for dispensing fluids, an ultrasonic horn for lysing cells or spores, and a proprietary I-CORE® thermocycler for performing real-time PCR and detection. The Xpert MRSA/SA Assay includes reagents for the simultaneous detection of the target organisms, SA and MRSA. The primers and probes in the Xpert MRSA/SA Assay detect nucleic acid sequences of the staphylococcal protein A (spa), the gene for MecA-Mediated Oxacillin resistance (mecA), and staphylococcal cassette chromosome (SCCmec) inserted in the SA chromosomal attB site. The test includes a sample processing control (SPC) to control for adequate processing of the target bacteria and to monitor the presence of inhibitor(s) in the PCR assay. The Probe Check Control (PCC) verifies reagent rehydration, PCR tube filling in the cartridge, probe integrity, and dye stability.

Here's a breakdown of the acceptance criteria and study details for the Cepheid Xpert MRSA/SA SSTI Assay, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state "acceptance criteria" numerical targets in a table format. Instead, it presents performance data (Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA)) in comparison to a reference culture method. These performance metrics implicitly serve as the acceptance criteria for regulatory approval.

Implicit Acceptance Criteria (Performance Targets) and Reported Device Performance

Note: The document states "The test results showed the Xpert MRSA/SA SSTI to be substantially equivalent to the current standard of care" and explicitly lists performance data, which are interpreted as meeting implicit acceptance criteria for regulatory clearance.

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

• Clinical Test Set Sample Size: A total of 848 specimens were collected from patients.
  • No Antibiotic Use (within 3 weeks): 441 subjects
  • Unknown Antibiotic Use (within 3 weeks): 200 subjects
  • Known Antibiotic Use (within 3 weeks): 207 subjects
• Data Provenance: The study was a multi-site prospective investigation conducted at four US institutions. This means the data was collected specifically for this study in a forward-looking manner, and from within the United States.

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

The document states that the reference culture method involved "confirmation of presumptive positive colonies was performed with catalase, tube coagulase, and Gram stain. MecA-Mediated Oxacillin resistance was tested by disk diffusion test using a 30 µg cefoxitin disk and cutoff of 21/22 mm." This implies standard microbiology laboratory procedures were followed by trained laboratory personnel.

• Number of Experts: Not explicitly stated as a specific number of individual "experts."
• Qualifications of Experts: Implied to be trained microbiologists or clinical laboratory technologists experienced in performing and interpreting standard microbiology culture, identification (catalase, coagulase, Gram stain), and susceptibility testing (cefoxitin disk diffusion). The reference culture was performed at a "centralized laboratory," suggesting specialized expertise.

4. Adjudication Method for the Test Set

The document does not describe a formal adjudication method (like 2+1 or 3+1 consensus) for the ground truth of the clinical test set. The "reference culture" is treated as the definitive ground truth, performed at a centralized laboratory following established protocols. Implicitly, any discrepancies would be resolved by the standard operating procedures of that centralized laboratory, but a multi-reader adjudication process is not detailed.

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

No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not done. This device is an automated in vitro diagnostic test (Nucleic Acid Amplification Test) performed on a GeneXpert Dx System, designed to provide results directly without requiring human interpretation of complex images or data that AI might otherwise augment for human readers. Its comparison is against a reference culture method.

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

Yes, a standalone performance study was done. The Xpert MRSA/SA SSTI Assay is an automated system (algorithm only without human interpretation in the loop beyond sample preparation and loading). The "Performance Characteristics" and "Clinical Performance" sections detail the direct comparison of the Xpert MRSA/SA SSTI Assay's results against the reference culture method, making it a standalone evaluation.

7. The Type of Ground Truth Used

The primary ground truth used for the clinical performance study was reference microbiology culture and susceptibility testing. This involved:

• Enrichment in trypticase soy broth.
• Streaking onto plates with and without cefoxitin.
• Sub-culturing presumptive colonies onto blood agar.
• Confirmation of positive colonies with catalase, tube coagulase, and Gram stain for Staphylococcus aureus (SA) identification.
• MecA-Mediated Oxacillin resistance tested by disk diffusion using a 30 µg cefoxitin disk and a cutoff of 21/22 mm for MRSA identification.

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 (AI). This device is a real-time PCR assay, which is a molecular diagnostic method based on known biological reactions, not typically developed using machine learning algorithms that require distinct training sets.

However, if "training set" is interpreted more broadly as data used for analytical development and optimization, the document mentions extensive analytical studies:

• Analytical Inclusivity Study on CDC Staphylococcus aureus Specimens: 21 MRSA strains and 3 MSSA strains (total 24 strains).
• Analytical Inclusivity Study on Expanded Panel of Staphylococcus aureus Specimens: 121 additional Staphylococcus aureus strains (78 MRSA, 43 SA).
• Evaluation of Empty Cassette Variants: 22 isolates.
• Limit of Detection Studies: Multiple individual isolates (6 MRSA, 3 SA) each tested with 20 replicates at various concentrations.
• Linearity Studies: SA and MRSA isolates tested across ten-fold serial dilutions.
• Cross-reactivity Study: 105 strains (various bacteria and yeast).
• Evaluation of BORSA Strains: 7 BORSA strains.

These analytical tests provide the data and parameters for the assay's design and demonstrate its performance characteristics, which is analogous to the role of a training set for algorithm development, even though it's a traditional molecular test.

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

For the analytical studies (which can be considered analogous to a training/development phase for a molecular assay), the ground truth was established through:

• Confirmed Identification: Using CDC-reported strains, phylogenetically characterized strains, ATCC cultures, and NARSA strains.
• Phenotypic Testing: Catalase, tube coagulase, Gram stain, and cefoxitin disk diffusion (for oxacillin resistance) were used to characterize discrepant results and confirm the identity and resistance profile of isolates.
• Genetic Characterization: PFGE for USA types and SCCmec typing for MRSA strains provided detailed genetic ground truth.
• Known Concentrations: For LoD and linearity studies, bacterial concentrations were carefully quantified (CFU/sample) to establish precise analytical ground truth.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of the antimicrobial susceptibility of minute in the for Enterobacteriaceae and Non-Enterobacteriaceae Inculture analone bacteria isolates from pure culture belonging to the genera Staphylococcus, Enterococcus and Streptococcus.

This premarket notification is for cefazolin at concentrations of 0.5-32 µg/mL to Gram-negative ID/AST or AST only Phoenix panels with the removal of the limitations for Klebsiella oxytoca and Proteus mirabilis.

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. The system includes the following components:

• BD Phoenix instrument and software.
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents for AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum.
• BD Phoenix AST Broth used for performing AST tests only.
• BD Phoenix AST Indicator solution added to the AST Broth to aid in bacterial growth determination.

The Phoenix AST method is a broth based microdilution test. The Phoenix system utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurements of changes to the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and preliminarily identified as a Gram-negative or Gram-positive isolate. Phoenix panels are inoculated with a specified organism density and placed in the instrument.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35°C. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, or R (sensitive, intermediate, or resistant).

This document describes the validation of the BD Phoenix™ Automated Microbiology System for Cefazolin (0.5-32 µg/mL) against Gram-negative bacteria, focusing on the removal of limitations for Klebsiella oxytoca and Proteus mirabilis.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are not explicitly stated with numerical targets for Essential Agreement (EA) and Category Agreement (CA) in the provided text. However, the study aims to demonstrate substantially equivalent performance when compared with the CLSI reference broth microdilution method, as defined in the FDA guidance "Class II Special Controls Guidance Document: Antimicrobial Susceptibility Test (AST) Systems".

Based on common FDA guidance for AST systems, "substantial equivalence" often implies that the new system's performance (EA and CA) is non-inferior to and ideally very similar to the reference method. While specific thresholds are not provided in this excerpt, typical FDA expectations for AST devices often include:

• Essential Agreement (EA): Generally ≥ 90%
• Category Agreement (CA): Generally ≥ 90%
• Minor Errors (mE): Typically ≤ 7-10%
• Major Errors (ME): Typically ≤ 1.5-3%
• Very Major Errors (VME): Typically ≤ 1.5-3%

The provided Table 1 is incomplete and does not contain the actual performance data for Cefazolin. It is presented as:

Therefore, without the actual numerical data from Table 1, the device's reported performance against the acceptance criteria cannot be fully extracted from the provided text. The text does state that "The performance of the Phoenix System was assessed by calculating Essential Agreement (EA) and Category Agreement (CA) to expected/reference results for all isolates tested." and "The data collected from the substantial equivalence studies demonstrate that testing on the BD Phoenix™ Automated Microbiology System with this antimicrobial agent is substantially equivalent as outlined in the FDA guidance document". This implies the device met the unstated acceptance criteria for substantial equivalence.

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

• Sample Size: The text mentions "Clinical, stock and challenge isolates were tested." No specific total number of isolates for the overall clinical study is provided. However, Table 1 (incomplete) shows a column with "197" which might refer to the number of isolates or a subset of them.
• Data Provenance: The data was collected from "multiple geographically diverse sites across the United States." The study included "Clinical, stock and challenge isolates." Clinical isolates are typically prospective or retrospectively collected real-world samples, while stock and challenge isolates are curated collections often used for specific performance assessments.

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

The ground truth for the test set was established by the CLSI (formerly NCCLS) reference broth microdilution method. This is an established standard laboratory procedure, not an opinion-based expert consensus. Therefore, the concept of "number of experts" and their "qualifications" in the traditional sense of subjective interpretation (e.g., radiologists reading images) does not directly apply here. The "experts" would be the trained laboratory personnel who meticulously performed the CLSI reference method.

4. Adjudication Method for the Test Set

Not applicable. The ground truth was established by the CLSI reference broth microdilution method, which is an objective measurement, not a subjective interpretation requiring adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No. This study is an evaluation of an automated AST system against a reference method, not a multi-reader multi-case study involving human readers with/without AI assistance. The device itself is an automated system for antimicrobial susceptibility testing.

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

Yes, this was a standalone performance study. The BD Phoenix™ Automated Microbiology System is an automated system, and its results were compared directly to the CLSI reference method. The device's performance is reported as its direct output (MIC values and categories S, I, R).

7. The Type of Ground Truth Used

The ground truth used was the CLSI reference broth microdilution method. This is a laboratory standard considered the gold standard for determining minimum inhibitory concentrations (MICs) and antimicrobial susceptibility categories. It represents a precise, objective, and well-established scientific method.

8. The Sample Size for the Training Set

The document does not explicitly mention a "training set" in the context of machine learning or algorithm development. The study described focuses on the validation of the device's performance against reference methods. Automated microbiology systems typically use predefined algorithms and pre-calibrated reagents based on extensive prior development and internal testing, rather than a "training set" in the typical AI sense during the final validation for substantial equivalence submission. If there was an internal development phase, the sample size for that "training" would not be disclosed in this type of regulatory summary.

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

As no specific "training set" is mentioned in the context of this regulatory submission, this question cannot be answered from the provided text. The performance evaluation is against the CLSI reference broth microdilution method as the ground truth.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic gram-negative and gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most Gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaceae and Non-Enteriaceae and most Gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus, Enterococcus, and Streptococcus.

This premarket notification is for the addition of the antimicrobial agent ertapenem at concentrations of 0.25-32 ug/mL to Gram-negative ID/AST or AST only Phoenix panels. Erapenem has been shown to be active in vitro against most strains of microorganisms listed below, as described in the FDA-approved package insert for this antimicrobial agent.

Active In Vitro and in Clinical Infections Against:
Escherichia coli
Klebsiella pneumoniae

Active In Vitro:
Citrobacter freundii
Citrobacter koseri
Enterobacter aerogenes
Enterobacter cloacae
Klebsiella oxytoca
Morganella morganii
Proteus mirabilis
Proteus vulgaris
Providencia rettgeri
Providencia stuartii
Serratia marcescens

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. The system includes the following components:

• BD Phoenix instrument and software. .
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents . for AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum. .
• BD Phoenix AST Broth used for performing AST tests only. .
• BD Phoenix AST Indicator solution added to the AST Broth to aid in bacterial growth . determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and preliminarily identified as a Gram-negative or Gram-positive isolate. Phoenix panels are inoculated with a specified organism density and placed into the instrument.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurements of changes to the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35°C. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S. I, or R (sensitive, intermediate, or resistant).

Here's a breakdown of the acceptance criteria and study information based on the provided text, adapted as if it were for an AI device for clarity in a modern context.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size: The document indicates that "Clinical, stock and challenge isolates were tested." However, specific numerical sample sizes for each type of isolate are not provided in the summary.
• Data Provenance: The data was collected from "multiple geographically diverse sites across the United States." The study used a combination of:
  • Clinical isolates: Presumably prospective as they represent real-world samples.
  • Stock isolates: Likely retrospective or pre-existing lab strains.
  • Challenge isolates: Likely retrospective or pre-existing, used for specific performance verification against known results.

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

The document does not mention the use of experts to establish ground truth for the test set. Instead, the ground truth was established by:

• CLSI reference broth microdilution method: This is a standardized, established laboratory method, not reliant on individual expert interpretation for each case.
• Expected results for Challenge set isolates: These "expected results" would be pre-determined values for known strains, which could have been established by consensus or by previous rigorous testing using reference methods.

4. Adjudication Method for the Test Set

No explicit adjudication method (e.g., 2+1, 3+1) is described for the test set. The comparison was made directly against the CLSI reference broth microdilution method or expected results for challenge isolates, which serve as the definitive standard.

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

• No, an MRMC comparative effectiveness study was not done. This study evaluates the performance of an automated microbiology system (BD Phoenix™) against a reference standard (CLSI broth microdilution method) for antimicrobial susceptibility testing. It does not involve human readers or assess the improvement of human readers with AI assistance.

6. Standalone (Algorithm Only) Performance Study

• Yes, a standalone performance study was done. The entire study is a standalone evaluation of the BD Phoenix™ Automated Microbiology System (which incorporates an algorithm to interpret results) by comparing its outputs (MIC values and category interpretations) against the CLSI reference method. There is no human-in-the-loop component described in the performance evaluation.

7. Type of Ground Truth Used

The primary ground truth used was the CLSI reference broth microdilution method (AST panels prepared according to NCCLS M7). For challenge isolates, "expected results" were used.

8. Sample Size for the Training Set

The document does not provide information on the sample size used for the training set for the BD Phoenix™ system's underlying algorithms. This is typical for such submissions, as the focus is on the performance of the final, already-developed device.

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

The document does not provide information on how the ground truth for the training set was established. Similar to the training set sample size, this information is not typically included in the summary for a 510(k) submission for a device like this, which is an automated system rather than a de novo AI model with publicly disclosed training data. It's assumed the system was developed and validated internally using appropriate microbiological standards.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most Gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaceae and Non-Enterobacteriaceae and most Gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus , Enterococcus, and Streptococcus.

This premarket notification is for the addition of the antimicrobial agent moxifloxacin at concentrations of 0.0625-8 ug/mL to Streptococcus ID/AST or AST only Phoenix panels. Moxifloxacin has been shown to be active in vitro against most strains of microorganisms listed below, as described in the FDA-approved package insert for this antimicrobial agent.

Active In Vitro and in Clinical Infections Against:

Streptococcus anginosus Streptococcus constellatus Streptococcus pneumoniae (including multi-drug resistant strains [MDRSP]) Streptococcus pyogenes

Active In Vitro Against:

Streptococcus agalactiae Streptococcus viridans group

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. For testing Streptococcus species the system includes the following components:

• . BD Phoenix instrument and software.
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents . or AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum. .
• . BD Phoenix AST-S Broth used for performing AST tests only.
• BD Phoenix AST-S Indicator solution added to the AST Broth to aid in bacterial growth . determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene trav with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and preliminarily identified as a Gram-negative or Gram-positive isolate. For each isolation equivalent to a 0.5 McFarland standard is prepared in Phoenix ID broth.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurements of changes to the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35°C. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, R or N (susceptible, intermediate, resistant or not susceptible).

The acceptance criteria and study proving the device meets these criteria for the BD Phoenix™ Automated Microbiology System for moxifloxacin (0.0625-8 µg/mL) with Streptococcus ID/AST or AST only Phoenix panels are described below:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Clinical Studies: The document mentions "Clinical, stock and challenge isolates" were tested.
  • Clinical Isolates: "all isolates tested" and tables showing "1050" isolates for Accuracy and Performance with Moxifloxacin. These were compared to the CLSI reference broth microdilution method.
  • Challenge Isolates: Compared to expected results. The number of challenge isolates is not explicitly stated separately but is implied to be part of the "1050" total if not specified otherwise.
• Reproducibility Studies: "a panel of streptococcal isolates" was used. The exact number is not explicitly stated.
• Data Provenance: "multiple geographically diverse sites across the United States" (for clinical studies). The study type is explicitly a "clinical study" comparing the device to a reference method, which implies prospective data collection for this purpose, though the source of the isolates (e.g., retrospective collection of patient samples) is not detailed.

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

The document does not explicitly state the number of experts or their qualifications for establishing the ground truth. The ground truth was established by:

• CLSI reference broth microdilution method: This method itself is the gold standard for antimicrobial susceptibility testing, implying adherence to a standardized protocol rather than individual expert interpretation for each case.
• Expected results for Challenge isolates: These are typically well-characterized strains with known susceptibility profiles.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method involving multiple experts for discrepant results. Instead, it relies on quantitative comparison metrics (Essential Agreement and Category Agreement) against the CLSI reference method or expected results for challenge isolates. The system determines an MIC value, which is then compared rather than subjective interpretation requiring adjudication.

5. If a Multi-reader Multi-case (MRMC) Comparative Effectiveness Study Was Done, If So, What Was the Effect Size of How Much Human Readers Improve with AI vs Without AI Assistance

This is not applicable. The BD Phoenix™ Automated Microbiology System is an automated system for identification and antimicrobial susceptibility testing. It is a standalone device that performs these tests without direct human "reading" of the results in the traditional sense or human-in-the-loop performance improvement as seen in AI-assisted diagnostics. The system provides MIC values and categorical interpretations automatically.

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

Yes, a standalone performance study was done. The entire study evaluates the performance of the BD Phoenix™ Automated Microbiology System (an automated device/algorithm) in isolation, comparing its output directly to the CLSI reference broth microdilution method. The system itself determines the MIC values and interpretations.

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

The ground truth used was primarily the CLSI reference broth microdilution method for clinical isolates. For challenge isolates, the ground truth was "expected results," which refers to the known susceptibility profile of those well-characterized strains. This is a recognized "gold standard" laboratory method.

8. The Sample Size for the Training Set

The document does not provide information regarding a distinct training set sample size. This type of regulatory submission typically focuses on validation and performance testing rather than the details of algorithm development and training, as the "algorithm" here relates to the system's ability to interpret growth patterns rather than a machine learning model developed with a large labeled dataset in the modern sense. The system's "learning" or calibration would have been done during its development, but these specifics are not part of this 510(k) summary.

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

Given that no specific training set is detailed, the method for establishing its ground truth is also not described. As mentioned in point 8, the submission focuses on the validation of the final device against reference methods rather than the internal development details.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most Gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaceae and Non-Enterobacteriaceae and most Gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus . Enterococcus, and Streptococcus.

This premarket notification is for the addition of the antimicrobial agent cefuroxime at concentrations of 0.125-4 µg/mL to Streptococcus ID/AST or AST only Phoenix panels. Cefuroxime has been shown to be active in vitro against most strains of microorganisms listed below, as described in the FDA-approved package insert for this antimicrobial agent.

Active In Vitro and in Clinical Infections Against:

Streptococcus pneumoniae Streptococcus pyogenes (and other streptococci)

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. For testing Streptococcus species the system includes the following components:

• . BD Phoenix instrument and software.
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents . or AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum. .
• BD Phoenix AST-S Broth used for performing AST tests only. .
• BD Phoenix AST-S Indicator solution added to the AST Broth to aid in bacterial growth . determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells The I nooms parel to a couler anisms for susceptibility testing must be a pure culture and preliminarily identified as a Gram-negative or Gram-positive isolate. For each isolate, an inoculation equivalent to a 0.5 McFarland standard is prepared in Phoenix ID broth.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. indicator for the detection of erganism go as well as bacterial turbidity are used in the determination Mrasurements of enanges to T panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35℃. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, R or N (susceptible, intermediate, resistant or not susceptible).

The provided text describes the 510(k) summary for the BD Phoenix™ Automated Microbiology System for use with the antimicrobial agent cefuroxime. Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implied by the statement that the system has demonstrated "substantially equivalent performance" when compared to the CLSI reference broth microdilution method and the FDA Draft guidance document. Specifically, the performance was assessed by Essential Agreement (EA) and Category Agreement (CA). While explicit numerical acceptance criteria for EA and CA are not stated, the study concludes substantial equivalence, indicating these metrics met the required thresholds.

Note: The provided Table 1 is severely corrupted with non-sensical text and numbers, making it impossible to extract the specific numerical values for Essential Agreement and Category Agreement. However, the text explicitly states the conclusions that the studies demonstrated substantial equivalence.

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

• Test Set Sample Size: The exact number of clinical, stock, and challenge isolates tested is not explicitly stated in the provided text. It mentions "Clinical, stock and challenge isolates were tested."
• Data Provenance: The isolates were tested "across multiple geographically diverse sites across the United States." This indicates prospective data collection for the clinical studies.

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

This information is not provided in the document. The reference method for comparison (CLSI reference broth microdilution method) is mentioned, but details on the human component of ground truth determination are absent.

4. Adjudication Method for the Test Set

This information is not provided in the document. The performance was assessed by comparing Phoenix System results to CLSI reference method results. The process for resolving discrepancies or establishing the "ground truth" for the reference method itself is not detailed.

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

A MRMC comparative effectiveness study was not performed as described in the typical sense for imaging devices. This device is an automated microbiology system. The comparison was between the automated system's output and a reference laboratory method (CLSI broth microdilution), not between human readers with and without AI assistance.

6. Standalone Performance

Yes, a standalone performance evaluation was done. The entire study describes the performance of the algorithm only (BD Phoenix™ Automated Microbiology System) in determining antimicrobial susceptibility without a human-in-the-loop for interpretive decisions once the system generates results. The system automatically takes readings and interprets them to give MIC values and category interpretations.

7. Type of Ground Truth Used

The ground truth used for the clinical studies was the CLSI reference broth microdilution method. For challenge isolates, results were compared "to the expected results," implying a predefined or known ground truth for those specific strains.

8. Sample Size for the Training Set

The document does not explicitly mention a separate "training set" or its sample size. The description focuses on the evaluation (test) phase to demonstrate substantial equivalence, rather than a development or training phase for the algorithm.

9. How Ground Truth for the Training Set Was Established

As no explicit training set is described, information on how its ground truth was established is not available in the provided text.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

This premarket notification is for the addition of the antimicrobial agent meropenem at concentrations of 0.0313-2 µg/mL to Streptococcus ID/AST or AST only Phoenix panels. Meropenem has been shown to be active in vitro against most strains of microorganisms listed below, as described in the FDA-approved package insert for this antimicrobial agent.

Active In Vitro and in Clinical Infections Against:

Streptococcus pneumoniae (excluding penicillin-resistant strains) Viridans group streptococci

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. For testing Streptococcus species the system includes the following components:

• BD Phoenix instrument and software. .
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents . or AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum. .
• BD Phoenix AST-S Broth used for performing AST tests only. .
• BD Phoenix AST-S Indicator solution added to the AST Broth to aid in bacterial growth . determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and containing driver organisme or Gram-negative or Gram-positive isolate. For each isolate, an inoculation equivalent to a 0.5 McFarland standard is prepared in Phoenix ID broth.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Indicator for the deterative of the indicator as well as bacterial turbidity are used in the determination of bacterial growth. Each AST panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35℃. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, R or N (susceptible, intermediate, resistant or not susceptible).

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

1. Table of Acceptance Criteria and Reported Device Performance

The device is an Antimicrobial Susceptibility Test (AST) system. The acceptance criteria are based on Essential Agreement (EA) and Category Agreement (CA) with a CLSI reference broth microdilution method.

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

• Sample Size for Test Set (Clinical Studies): The table for Essential Agreement (EA) and Category Agreement (CA) shows n=1935 for the "Antimicrobial" (Meropenem) for streptococcal organisms. This represents the number of isolates tested.
• Data Provenance: "Clinical, stock and challenge isolates were tested across multiple geographically diverse sites." This indicates a mix of retrospective (stock and challenge) and prospective (clinical) data from various geographical locations.

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

The document does not specify the number or qualifications of experts used to establish the ground truth. The ground truth method, CLSI reference broth microdilution, is a standardized laboratory procedure, not typically relying on expert consultation for each case's outcome.

4. Adjudication Method for the Test Set

The document does not describe an adjudication method for disagreements. The comparison is made against a standardized "CLSI reference broth microdilution method." Discrepancies would likely be investigated for technical or procedural errors, rather than expert adjudication in the traditional sense of medical image interpretation.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and 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 study focuses on the performance of an automated microbiology system (BD Phoenix) against a reference standard (CLSI broth microdilution), not on human reader performance with or without AI assistance. The device is a standalone diagnostic tool, not an AI-assisted diagnostic aid for human readers in the context of these types of studies.

6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, a standalone performance study was done. The entire study describes the performance of the "BD Phoenix Automate Microbiology System" (the device/algorithm) in isolation, compared to the CLSI reference method. The results (EA, CA) are for the device's output without direct human interpretation influencing the final MIC or category.

7. The Type of Ground Truth Used

The type of ground truth used is the CLSI reference broth microdilution method. This is a laboratory-based, standardized, and highly accepted method for determining antimicrobial susceptibility.

8. The Sample Size for the Training Set

The document does not provide information on the training set size. This is common for this type of 510(k) submission, as the focus is on the validation performance of the final device against a reference method. The device's underlying "algorithm" (how it interprets readings) would have been developed and "trained" prior to this validation.

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

The document does not specify how the ground truth for the training set was established, nor does it describe the training process. This information is typically proprietary to the manufacturer and not required for 510(k) submissions focusing on validation against established standards. However, it can be inferred that similar CLSI reference methods or other established laboratory techniques would have been used during the development and calibration of the system.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic Gram-negative and Gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most Gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaeae and Non-Enterobacteriaeae and Gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus and Enterococcus.

This premarket notification is for the addition of the antimicrobial agent Cefazolin at concentrations of 2-32 µg/mL to Gram Positive ID/AST or AST only Phoenix panels. Cefazolin has been shown to be active in vitro and in clinical infections against: Staphylococcus aureus (including penicillinase-producing strains) and Staphylococcus epidermidis.

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the rapid identification (ID) and antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. The system includes the following components:

• BD Phoenix instrument and software.
• BD Phoenix panels for organism ID testing and antimicrobial agents for AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum.
• BD Phoenix AST Broth used for performing AST tests only.
• BD Phoenix AST Indicator solution added to the AST Broth to aid in bacterial growth determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and preliminary identification as a Gram-negative or Gram-positive isolate. For each isolate, an inoculation suspension equivalent to a 0.5 McFarland standard is prepared in Phoenix ID broth.

The Phoenix AST method is a broth based microdilution test. The Phoenix system utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurements of changes to the indicator as well as bacterial turbidity are used in the determination of organism growth. The Phoenix panels contain various antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35 degrees Celsius. The instrument takes readings of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, or R (sensitive, intermediate, or resistant).

Acceptance Criteria and Study Details for BD Phoenix™ Automated Microbiology System - Cefazolin 2-32 µg/mL

1. Table of Acceptance Criteria and Reported Device Performance

The provided document defines the acceptance criteria in terms of "Essential Agreement (EA)" and "Category Agreement (CA)" with the NCCLS reference broth microdilution method. The performance is summarized in Table 1.

Note: While specific numeric acceptance criteria (e.g., "EA > 95%") are not explicitly listed in the provided text, the conclusion that the device "demonstrates substantially equivalent performance" based on the observed 100% EA and CA strongly implies that these results met the internal or regulatory acceptance thresholds for substantial equivalence under the FDA Draft guidance document, "Guidance on Review Criteria for Assessment of Antimicrobial Susceptibility Devices," March 8, 2000.

2. Sample size used for the test set and the data provenance

• Sample Size for Test Set: 507 isolates (n=507) of Gram-positive organisms.
• Data Provenance:
  • Country of Origin: United States. The study states, "Clinical, stock and challenge isolates were tested across multiple geographically diverse sites across the United States."
  • Retrospective or Prospective: Not explicitly stated as retrospective or prospective for the clinical isolates. However, the testing of "stock and challenge isolates" alongside "clinical isolates" suggests a methodology that combines existing collections (stock/challenge, which could be considered retrospective or curated) with potentially newly acquired clinical isolates (prospective or concurrent). The overall context of a premarket study for a new antimicrobial agent addition to an existing system often involves a mix to ensure comprehensive evaluation.

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

The document does not specify the number of experts or their qualifications. The "ground truth" (reference results) was established using the NCCLS reference broth microdilution method. This method is a standardized laboratory procedure, not typically dependent on expert interpretation for establishing the result itself, but rather on adherence to the protocol by trained laboratory personnel.

4. Adjudication method for the test set

The document does not describe an adjudication method involving multiple human readers or experts for the test set. The comparison was made between the BD Phoenix System results and the results obtained from the NCCLS reference broth microdilution method, which is a direct, standardized laboratory assay.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No, a multi-reader, multi-case (MRMC) comparative effectiveness study was not done.

This device (BD Phoenix Automated Microbiology System) is an automated system for antimicrobial susceptibility testing, which provides automated results (MIC values and categorical interpretations). It does not involve human "readers" interpreting images or clinical data where AI assistance would be applicable in the typical sense of an MRMC study for imaging or clinical decision support. The system is the "AI" (or automated algorithm) that performs the test and provides the interpretation.

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

Yes, a standalone performance study was done.

The entire study described is a standalone performance evaluation of the "BD Phoenix™ Automated Microbiology System" (the device/algorithm) against the reference method. The system inherently operates as an "algorithm only" in generating the MIC and susceptibility category interpretations. There is no mention of a human-in-the-loop component being evaluated or integrated into the performance claims for this specific study; the device itself provides the final result.

7. The type of ground truth used

The type of ground truth used was NCCLS reference broth microdilution method. This is a laboratory reference standard (gold standard biological assay) for determining antimicrobial susceptibility, widely accepted in microbiology.

8. The sample size for the training set

The document does not specify a separate "training set" or its sample size. The description focuses on the validation of the Cefazolin agent on the existing BD Phoenix system. Automated microbiology systems typically rely on pre-established algorithms and decision rules derived from extensive historical data and expert panels during their initial development. The current submission is for an addition of an antimicrobial agent to an already established system, so the performance evaluation is focused on the new agent's accuracy with the existing system, rather than a de novo algorithm training.

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

As no separate "training set" is explicitly mentioned for this specific submission, the method for establishing its ground truth is not provided in the document. For the development of the core BD Phoenix system itself, ground truth for algorithm training would have historically been established through extensive testing against reference methods like NCCLS broth microdilution, often corroborated with expert microbiological interpretation over many years of data collection.

Ask a specific question about this device

The BD Phoenix™ Automated Microbiology System is intended for the rapid identification and in vitro antimicrobial susceptibility testing of isolates from pure culture of most aerobic and facultative anaerobic gram-negative and gram-positive bacteria of human origin.

The BD Phoenix™ Automated Microbiology System is intended for in vitro quantitative determination of antimicrobial susceptibility by minimal inhibitory concentration (MIC) of most gram-negative aerobic and facultative anaerobic bacteria isolates from pure culture for Enterobacteriaceae and Non-Enterobacteriaceae and most gram-positive bacteria isolates from pure culture belonging to the genera Staphylococcus and Enterococcus.

This premarket notification is for the addition of the antimicrobial agent ceftazidime at concentrations of 0.5-64 µg/mL to gram-negative ID/AST or AST only Phoenix panels. Ceftazidime has been shown to be active in vitro against most strains of microorganisms listed below, as described in the FDA-approved package insert for this antimicrobial agent.

Active In Vitro and in Clinical Infections Against:
Citrobacter spp., excluding Citrobacter freundii
Enterobacter spp., including Enterobacter cloacae, and Enterobacter aerogenes
Escherichia coli
Klebsiella spp., including Klebsiella pneumoniae
Proteus mirabilis
Proteus vulgaris
Pseudomonas spp., including Pseudomonas aeruginosa
Serratia spp.

The BD Phoenix Automated Microbiology System (Phoenix System) is an automated system for the in vitro diagnostic antimicrobial susceptibility testing (AST) of clinically relevant bacterial isolates. The system includes the following components:

• BD Phoenix instrument and software.
• BD Phoenix panels containing biochemicals for organism ID testing and antimicrobial agents for AST determinations.
• BD Phoenix ID Broth used for performing ID tests and preparing AST Broth inoculum.
• BD Phoenix AST Broth used for performing AST tests only.
• BD Phoenix AST Indicator solution added to the AST Broth to aid in bacterial growth determination.

The Phoenix panel is a sealed and self-inoculating molded polystyrene tray with 136 micro-wells containing dried reagents. Organisms for susceptibility testing must be a pure culture and equivalent to a 0.5 McFarland standard is prepared in Phoenix ID Broth.

The Phoenix AST method is a broth based microdilution test. The Phoenix System utilizes a redox indicator for the detection of organism growth in the presence of an antimicrobial agent. Measurements of changes to the indicator as bacterial turbidity are used in the determination of MIC values. Each Phoenix panel configuration contains several antimicrobial agents with a wide range of two-fold doubling dilution concentrations.

The instrument houses the panels where they are continuously incubated at a nominal temperature of 35 degrees C. The instrument monitors the growth in each well of the panels every 20 minutes. The readings are interpreted to give an identification of the isolate, minimum inhibitory concentration (MIC) values and category interpretations, S, I, or R (sensitive, intermediate, or resistant).

The provided text describes the performance of the BD Phoenix™ Automated Microbiology System when testing the antimicrobial agent Ceftazidime against gram-negative organisms.

Here's an analysis of the acceptance criteria and study details:

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

The study refers to the FDA guidance document "Class II Special Controls Guidance Document: Antimicrobial Susceptibility Test (AST) Systems; Guidance for Industry and FDA," February 5, 2003, for establishing the acceptance criteria. The performance is assessed using Essential Agreement (EA) and Category Agreement (CA).

Note: While the exact numerical acceptance criteria for EA and CA are not explicitly stated in the provided text, they are generally established to be high (e.g., above 90%) for substantial equivalence in AST device submissions according to FDA guidance documents. The document states that performance was "substantially equivalent as outlined in the FDA guidance document."

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

• Test Set Sample Size:
  • Clinical Isolates: The table shows n = 1706 for Ceftazidime, which likely represents the total number of clinical isolates tested.
  • Reproducibility Studies: "one lot of gram-negative Phoenix panels containing this antimicrobial agent and associated reagents" was tested across three different days for intra- and inter-site reproducibility. The exact number of isolates used in these reproducibility studies is not explicitly stated, but they involved "gram-negative isolates."
• Data Provenance: "Clinical, stock and challenge isolates were tested across multiple geographically diverse sites across the United States." This indicates prospective and retrospective data (clinical isolates, stock and challenge isolates) from multiple locations within the United States.

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

The text does not specify the "number of experts" or their "qualifications." The ground truth was established by the NCCLS reference broth microdilution method, which is a standardized laboratory procedure rather than an expert consensus based on individual human interpretation.

4. Adjudication method for the test set:

Not applicable. The ground truth was based on the NCCLS reference broth microdilution method, which is a quantitative laboratory test. There was no need for human expert adjudication in the context of interpreting the reference method 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:

Not applicable. This device is an automated microbiology system and not an AI-assisted diagnostic tool that human readers would use. The comparison is between the automated system's performance and a standardized reference laboratory method, not human reader performance.

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

Yes, a standalone performance study was done. The entire study focuses on the performance of the BD Phoenix™ Automated Microbiology System (an automated device/algorithm) in determining antimicrobial susceptibility. Its results are compared directly to the NCCLS reference method without human interpretation being part of the Phoenix system's performance evaluation.

7. The type of ground truth used:

The ground truth used was the NCCLS reference broth microdilution method, which is a standardized, established laboratory test for antimicrobial susceptibility. This is a reference standard method rather than expert consensus, pathology, or outcomes data.

8. The sample size for the training set:

The document does not explicitly mention a "training set" or its size. This type of submission (510(k) for an AST system) typically focuses on the device's validation against a reference method rather than describing an AI model's training process. The system is likely based on established principles of bacterial growth and redox indicators, not a machine learning model that requires a discrete training set in the conventional sense.

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

Not applicable, as a distinct training set (in the machine learning sense) is not described or implied for this automated system. The device's underlying principles are based on known biological and chemical reactions for bacterial growth and metabolism.

Ask a specific question about this device