Search Results

PASCO MIC AND MIC/ID PANELS are used for quantitatively measuring (with the exception of the Breakpoint/ID panel which provides qualitative measurement or category results) the susceptibility of rapidly growing aerobic and facultative anaerobic bacterial pathogens to a battery of antimicrobial agents and determining the biochemical identification of those organisms.

This 510(k) notification is for the addition of Cefotaxime to Pasco panels at concentrations of 4-0.03 mcg/ml for use in determining the susceptibility of S. pneumoniae and non-pneumococcal streptococci.

Varying concentrations of antimicrobial agents (usually in two-fold dilutions) are dispensed into the Pasco panels and the panels are then frozen. Panels are thawed prior to use, inoculated with the test organisms, incubated the traditional 16-24 hours and panels are then observed for visible growth or color changes as described in the package insert. The lowest concentration of each antimicrobial agent with no apparent visible growth of the test organism is recorded as the minimum inhibitory concentration (MIC). Changes in pH and production of specific metabolites from growth in biochemical substrates are interpreted as described in the package insert for conventional tubed media.

Here's a breakdown of the acceptance criteria and the study details for the Pasco MIC and MIC/ID Panels for Cefotaxime, based on the provided text:

Acceptance Criteria and Reported Device Performance

Note: The document refers to "acceptable" agreement levels without explicitly stating quantitative thresholds for these (e.g., "EA must be >90%"). However, the reported performance levels are presented as meeting these implicit criteria.

Study Details

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

   • Test Set Size:
     • 101 S. pneumoniae strains
     • 130 non-pneumococcal strains
     • 12 organisms for reproducibility testing
     • Unspecified number of CDC challenge strains (included within the 101 S. pneumoniae and 130 non-pneumococcal strains)
   • Data Provenance: The study was "performed at two sites using CDC challenge strains and clinical isolates." This indicates a combination of retrospective (CDC challenge strains) and prospective (clinical isolates) data. The country of origin for the clinical isolates is not specified, but Pasco is a US-based company, suggesting US clinical isolates.

2. 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 or qualifications of experts used. The ground truth ("reference panel") is likely established by a standardized method (e.g., broth microdilution) rather than expert consensus on individual results.

3. Adjudication method for the test set:

   • The document does not explicitly state an adjudication method. It mentions that one minor error in S. pneumoniae testing was "not considered resolved since the reference method MIC results did not change," implying that the reference method was the ultimate arbiter, rather than a separate adjudication process.

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. This study evaluates the performance of an Antimicrobial Susceptibility Test (AST) panel – essentially a laboratory diagnostic device. It does not involve AI or human readers in the context of image interpretation or other diagnostic tasks typically associated with MRMC studies. The device itself (the panel) is the "reader" in this context.

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

   • Yes, in a sense. The Pasco MIC and MIC/ID Panels are designed to provide quantitative or qualitative susceptibility results. While human observation is required to interpret visible growth or color changes, the "device performance" in terms of MIC determination is evaluated as a standalone entity compared to a reference method, not intertwined with human diagnostic variability. It's an in vitro diagnostic test, not an AI algorithm.

6. The type of ground truth used:

   • The ground truth was established using a "reference panel" (also referred to as the "reference method"). This implies a gold standard laboratory method for determining antimicrobial susceptibility, such as a broth microdilution method, which is the established gold standard for MIC determination.

7. The sample size for the training set:

   • The document does not explicitly state a separate "training set" size. The "test panels" were "prepared in-house at Pasco using routine manufacturing procedures." Antimicrobial susceptibility panels are typically subject to quality control during manufacturing and then evaluated in comparative studies like this, rather than going through a machine learning-style training/testing split. The data mentioned (101 S. pneumoniae strains, 130 non-pneumococcal strains) are used for the comparative testing, not for training an algorithm.

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

   • As there is no explicitly mentioned "training set" in the context of machine learning, this question is not directly applicable. If interpreting "training set" broadly as data used for developing the panel or its manufacturing process, the ground truth would still be established by standard microbiological reference methods and quality control procedures to ensure the accuracy of the antimicrobial agent concentrations and their performance.

Ask a specific question about this device