K212576 · Streck, Inc. · PMN · Jan 19, 2022 · Microbiology
Device Facts
Record ID
K212576
Device Name
MDx-Chex for BCID2
Applicant
Streck, Inc.
Product Code
PMN · Microbiology
Decision Date
Jan 19, 2022
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.3920
Device Class
Class 2
Indications for Use
MDx-Chex for BCID2 is intended for use as an external positive and negative assayed control to monitor the performance of the qualitative detection of yeast, Gram positive and Gram negative bacteria, as well as associated antimicrobial resistance genes, by the BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on FilmArray systems. Control 1 - GN: Gram negative bacteria: Acinetobacter colcoaceticus-baumannii complex, Bacteroides fragilis, Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella oxytoca, Klebsiella pneumoniae group, Proteus spp., Salmonella spp., Serratia marcescens, Haemophilus influenza, Neisseria meningitides, Pseudomonas aeruginosa, Stenotrophomonas maltophilia; antimicrobial resistance genes: KPC, CTX-M, IMP, NDM, OXA-48-like, VIM, mcr-1. Control 2 - GPY: Gram positive bacteria: Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunesis, Streptococcus agalactiae, Streptococcus pneumonia, Streptococcus pyogenes; yeast: Candida albicans, Candida auris, Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis, Cryptococcus neoformans/gatti; antimicrobial resistance genes: mecA/C and MREJ, vanA/B. This product is not intended to replace manufacturer controls provided with the device.
Device Story
MDx-Chex for BCID2 is a ready-to-use, liquid, full-process cellular control; contains inactivated microorganisms, human erythrocytes, leukocytes, and blood culture media components. Used in clinical laboratories to monitor the entire analytical process of the BioFire FilmArray BCID2 Panel, including sample lysis, nucleic acid isolation/purification, amplification, detection, and analysis. Acts as an external control to identify variations in the test system, such as PCR inhibitors or preanalytical variables. Operators process the control like a patient sample on the FilmArray system. Output is a qualitative result (detected/not detected) for specific pathogens and resistance genes. Benefits include ensuring test system reliability and identifying potential errors in the diagnostic workflow.
Clinical Evidence
No clinical data. Performance was established via bench testing, including multi-site reproducibility, internal precision, within-run testing, and lot-to-lot reproducibility studies. Across 240 samples in reproducibility studies, the device achieved 95.8% positive percent agreement (PPA) and 99.6% negative percent agreement (NPA). Stability studies (open-vial, closed-vial, and shipping) confirmed performance after 60+ days of storage and temperature stress. Matrix comparison studies confirmed the control performs identically to clinical blood culture samples.
Technological Characteristics
Ready-to-use liquid control; contains intact inactivated bacteria, human erythrocytes, leukocytes, and blood culture media components. Monitors lysis, nucleic acid isolation/purification, PCR inhibitor removal, amplification, and detection. Compatible with BioFire FilmArray Torch and 2.0 systems. Stability: 60 days at 2-25°C.
Indications for Use
Indicated for use as an external positive and negative assayed quality control material for the BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on FilmArray 2.0 and FilmArray Torch systems. Intended for clinical laboratory use to monitor the qualitative detection of Gram-positive bacteria, Gram-negative bacteria, yeast, and antimicrobial resistance genes in blood culture samples.
Regulatory Classification
Identification
An assayed quality control material for clinical microbiology assays is a device indicated for use in a test system to estimate test precision or to detect systematic analytical deviations that may arise from reagent or analytical instrument variation. This type of device consists of single or multiple microbiological analytes intended for use with either qualitative or quantitative assays.
Special Controls
An assayed quality control material for clinical microbiology assays must comply with the following special controls: (1) Premarket notification submissions must include detailed device description documentation and information concerning the composition of the quality control material, including, as appropriate: (i) Analyte concentration; Expected values: (ii) Analyte source: (iii) (iv) Base matrix; (v) Added components; (vi) Safety and handling information; and, (vii) Detailed instructions for use. (2) Premarket notification submissions must include detailed documentation, including line data as well as detailed study protocols and a statistical analysis plan used to establish performance, including: (i) Description of the process for value assignment and validation. (ii) Description of the protocol(s) used to establish stability. (iii) Line data establishing precision/reproducibility. (iv) Where applicable, assessment of matrix effects and any significant differences between the quality control material and typical patient samples in terms of conditions known to cause analytical error or affect assay performance. (v) Where applicable, identify or define traceability or relationship to a domestic or international standard reference material and/or method. (vi) Where applicable, detailed documentation related to studies for surrogate controls. (3) Premarket notification submissions must include an adequate mitigation (e.g., realtime stability program) to the risk of false results due to potential modifications to the assays specified in the device's 21 CFR 809.10 compliant labeling. (4) Your 21 CFR 809.10 compliant labeling must include the following: (i) The intended use in your 21 CFR 809.10(a)(2) and 21 CFR 809.10(b)(2) compliant labeling must include the following: (A) Assayed control material analyte(s); (B) Whether the material is intended for quantitative or qualitative assays: (C) Stating if the material is a surrogate control; (D)The system(s), instrument(s), or test(s) for which the quality control material is intended. (ii) The intended use in your 21 CFR 809.10(a)(2) and 21 CFR 809.10(b)(2) compliant labeling must include the following statement: "This product is not intended to replace manufacturer controls provided with the device." (iii)A limiting statement that reads "Quality control materials should be used in accordance with local, state, federal regulations, and accreditation requirements."
*Classification.* Class II (special controls). The special controls for this device are:(1) Premarket notification submissions must include detailed device description documentation and information concerning the composition of the quality control material, including, as appropriate:
(i) Analyte concentration;
(ii) Expected values;
(iii) Analyte source;
(iv) Base matrix;
(v) Added components;
(vi) Safety and handling information; and
(vii) Detailed instructions for use.
(2) Premarket notification submissions must include detailed documentation, including line data as well as detailed study protocols and a statistical analysis plan used to establish performance, including:
(i) Description of the process for value assignment and validation.
(ii) Description of the protocol(s) used to establish stability.
(iii) Line data establishing precision/reproducibility.
(iv) Where applicable, assessment of matrix effects and any significant differences between the quality control material and typical patient samples in terms of conditions known to cause analytical error or affect assay performance.
(v) Where applicable, identify or define traceability or relationship to a domestic or international standard reference material and/or method.
(vi) Where applicable, detailed documentation related to studies for surrogate controls.
(3) Premarket notification submissions must include an adequate mitigation (e.g., real-time stability program) to the risk of false results due to potential modifications to the assays specified in the device's 21 CFR 809.10 compliant labeling.
(4) Your 21 CFR 809.10 compliant labeling must include the following:
(i) The intended use of your 21 CFR 809.10(a)(2) and (b)(2) compliant labeling must include the following:
(A) Assayed control material analyte(s);
(B) Whether the material is intended for quantitative or qualitative assays;
(C) Stating if the material is a surrogate control; and
(D) The system(s), instrument(s), or test(s) for which the quality control material is intended.
(ii) The intended use in your 21 CFR 809.10(a)(2) and (b)(2) compliant labeling must include the following statement: “This product is not intended to replace manufacturer controls provided with the device.”
(iii) A limiting statement that reads “Quality control materials should be used in accordance with local, state, federal regulations, and accreditation requirements.”
{0}
FDA U.S. FOOD & DRUG ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY ONLY
## I Background Information:
A 510(k) Number
K212576
B Applicant
Streck, Inc
C Proprietary and Established Names
MDx-Chex for BCID2
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| PMN | Class II | 21 CFR 866.3920 - Assayed Quality Control Material For Clinical Microbiology Assays | MI - Microbiology |
## II Submission/Device Overview:
A Purpose for Submission:
To obtain a substantive equivalence determination for the MDx-Chex for BCID2.
B Measurand:
Multi-analyte quality control materials
C Type of Test:
MDx-Chex for BCID2 is intended for in vitro diagnostic use as an external assayed quality control material to monitor the qualitative amplification, detection, and identification steps of the laboratory nucleic acid test BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on the FilmArray 2.0 and FilmArray systems. The BioFire BCID2 Panel on the FilmArray 2.0 or FilmArray Torch system detects multiple bacterial and yeast nucleic acids and select genetic
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
determinants of antimicrobial resistance from positive blood cultures. Control 1 - GN: Gram negative bacteria: Acinetobacter colcoaceticus-baumannii complex, Bacteroides fragilis, Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella oxytoca, Klebsiella pneumoniae group, Proteus spp., Salmonella spp., Serratia marcescens, Haemophilus influenza, Neisseria meningitides, Pseudomonas aeruginosa, Stenotrophomonas maltophilia; antimicrobial resistance genes: KPC, CTX-M, IMP, NDM, OXA-48-like, VIM, mcr-1. Control 2 - GPY: Gram positive bacteria: Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunesis, Streptococcus agalactiae, Streptococcus pneumonia, Streptococcus pyogenes; yeast: Candida albicans, Candida auris, Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis, Cryptococcus neoformans/gatti; antimicrobial resistance genes: mecA/C and MREJ, vanA/B.
## III Intended Use/Indications for Use:
### A Intended Use(s):
MDx-Chex for BCID2 is intended for use as an external positive and negative assayed control to monitor the performance of the qualitative detection of yeast, Gram positive and Gram negative bacteria, as well as associated antimicrobial resistance genes, by the BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on FilmArray systems. Control 1 - GN: Gram negative bacteria: Acinetobacter colcoaceticus-baumannii complex, Bacteroides fragilis, Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella oxytoca, Klebsiella pneumoniae group, Proteus spp., Salmonella spp., Serratia marcescens, Haemophilus influenza, Neisseria meningitides, Pseudomonas aeruginosa, Stenotrophomonas maltophilia; antimicrobial resistance genes: KPC, CTX-M, IMP, NDM, OXA-48-like, VIM, mcr-1. Control 2 - GPY: Gram positive bacteria: Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunesis, Streptococcus agalactiae, Streptococcus pneumonia, Streptococcus pyogenes; yeast: Candida albicans, Candida auris, Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis, Cryptococcus neoformans/gatti; antimicrobial resistance genes: mecA/C and MREJ, vanA/B. This product is not intended to replace manufacturer controls provided with the device.
### B Indication(s) for Use:
See Indications for Use below.
### C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
For in vitro diagnostic use only
### D Special Instrument Requirements:
The MDx-Chex for BCID2 was evaluated on the FilmArray 2.0 and FilmArray Torch instruments.
K212576 - Page 2 of 15
{2}
IV Device/System Characteristics:
# A Device Description:
Streck Molecular Control (SMC; trade name: MDx-Chex for BCID2) is an external positive and negative assayed control consisting of two levels of control, Control 1- Gram negative (GN) and Control 2- GPY (Gram positive and yeast). Each control contains stabilized human leukocytes and erythrocytes, and inactivated bacteria in a simulated blood culture media. Both controls contain target sequence in intact and inactivated bacterial and yeast cells, as well as antimicrobial gene targets (in relevant pathogens that are intact and inactivated) that are detected by the BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on FilmArray systems. MDx-Chex for BCID2 is a full-process, cellular-based control for the BioFire BCID2 Panel that evaluates the entire analytical process including sample lysis, nucleic acid isolation and purification, amplification, detection, and analysis, as well as the impact of PCR inhibitors and preanalytical variables. Each control for the MDx-Chex for BCID2 is processed separately according to the BioFire BCID2 Panel Package Insert for patient samples.
B Pathogens and antimicrobial resistance genes found in MDx-Chex for BCID2 and detected by BioFire BCID2 Panel assay
| Antimicrobial resistance genes | |
| --- | --- |
| CTX-M | mecA/C and MREJ (MRSA) |
| IMP | NDM |
| KPC | OXA-48-like |
| mcr-1 | vanA/B |
| mecA/C | VIM |
| Gram Positive Bacteria | |
| Enterococcus faecalis | Streptococcus spp. |
| Enterococcus faecium | Streptococcus agalactiae (Group B) |
| Listeria monocytogenes | Streptococcus pneumoniae |
| Staphylococcus spp. | Streptococcus pyogenes (Group A) |
| Staphylococcus aureus | |
| Staphylococcus epidermidis | |
| Staphylococcus lugdunensis | |
| Gram Negative Bacteria | |
| Acinetobacter calcoaceticus-baumannii complex | Enteric bacteria |
| Bacteroides fragilis | Enterobacter cloacae complex |
| Haemophilus influenzae | Escherichia coli |
| Neisseria meningitidis | Klebsiella aerogenes |
| Pseudomonas aeruginosa | Klebsiella oxytoca |
| Stenotrophomonas maltophilia | Klebsiella pneumoniae group |
| | Salmonella spp |
| | Serratia marcescens |
| Yeast | |
| Candida albicans | Candida parapsilosis |
| Candida auris | Candida tropicalis |
| Candida glabrata | Cryptococcus neoformans/gattii |
| Candida krusei | |
| Candida lusitaniae | C. neoformans/gattii |
| C. lusitaniae | C. neoformans/gattii |
| C. lusitaniae | C. neoformans/gattii |
| E. coli | E. coli |
| E. coli | E. coli |
| E. coli | E. coli |
K212576 - Page 3 of 15
{3}
C Principle of Operation:
Not Applicable
V Substantial Equivalence Information:
A Predicate Device Name(s):
FilmArray BCID2 Control Panel M416
B Predicate 510(k) Number(s):
K200010
C Comparison with Predicate(s):
| Device & Predicate Device(s): | K212576 | K200010 |
| --- | --- | --- |
| Device Trade Name | MDx-Chex for BCID2 | FilmArray BCID2 Control Panel M416 |
| General Device Characteristic Similarities | | |
| Intended Use/Indications For Use | MDx-Chex for BCID2 is intended for use as an external positive and negative assayed control to monitor the performance of the qualitative detection of yeast, Gram positive and Gram negative bacteria, as well as associated antimicrobial resistance genes, by the BioFire FilmArray Blood Culture Identification 2 (BCID2) Panel on FilmArray systems. Control 1 - GN: Gram negative bacteria: Acinetobacter colcoaceticus-baumannii complex, Bacteroides fragilis, Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella oxytoca, Klebsiella pneumoniae group, Proteus spp., Salmonella spp., Serratia marcescens, Haemophilus influenza, Neisseria meningitides, Pseudomonas aeruginosa, Stenotrophomonas maltophilia; antimicrobial resistance genes: KPC, CTX-M, IMP, NDM, OXA-48-like, VIM, mcr-1. Control 2 - GPY: Gram positive bacteria: Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, | FilmArray BCID2 Control Panel M416 is intended for use as an external positive and negative assayed quality control to monitor the performance of in vitro laboratory nucleic acid testing procedures for the qualitative detection of antimicrobial resistance genes: CTX-M, IMP, KPC, mcr-1, mecA/C, mecA/C and MREJ (MRSA), NDM, OXA-48-like, vanA/B, VIM; Gram positive and Gram-negative bacteria: Enterococcus faecalis, Enterococcus faecium, Listeria monocytogenes, Staphylococcus spp., Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunensis, Streptococcus spp., Streptococcus agalactiae (Group B), Streptococcus pneumoniae, Streptococcus pyogenes (Group A), Acinetobacter calcoaceticus baumannii complex, Bacteroides fragilis, Enteric bacteria, Enterobacter cloacae complex, Escherichia coli, Klebsiella aerogenes, Klebsiella oxytoca, |
K212576 - Page 4 of 15
{4}
| | Staphylococcus aureus, Staphylococcus epidermidis, Staphylococcus lugdunesis, Streptococcus agalactiae, Streptococcus pneumonia, Streptococcus pyogenes; yeast: Candida albicans, Candida auris, Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis, Cryptococcus neoformans/gatti; antimicrobial resistance genes: mecA/C and MREJ, vanA/B. This product is not intended to replace manufacturer controls provided with the device. | Klebsiella pneumoniae group, Proteus spp., Salmonella spp., Serratia marcescens, Haemophilus influenzae, Neisseria meningitidis, Pseudomonas aeruginosa and Stenotrophomonas maltophilia; and yeast pathogens: Candida albicans, Candida auris, Candida glabrata, Candida krusei, Candida parapsilosis, Candida tropicalis, and Cryptococcus neoformans/gattii on the BioFire Blood Culture Identification 2 (BCID2) Panel assay on FilmArray systems. FilmArray BCID2 Control Panel M416 is composed of synthetic DNA specifically designed for and intended to be used solely with the BioFire BCID2 Panel assay. This product is not intended to replace manufacturer controls provided with the device. |
| --- | --- | --- |
| Physical Format | Ready-to-Use Liquid | Same |
| Direction for Use | Process like patient sample | Same |
| Number of targets monitored in one assay | Multiple, >30 targets | Same |
| General Device Characteristic Differences | | |
| Composition | Intact inactivated bacteria, human erythrocytes and leukocytes, and relevant components of blood culture media | Synthetic DNA |
| Assay Steps Monitored | Lysis, nucleic acid isolation/purification/PCR inhibitor removal, amplification, detection, identification/data reporting | Reverse transcription, amplification, detection |
VI Standards/Guidance Documents Referenced:
None Referenced
VII Performance Characteristics (if/when applicable):
A Analytical Performance:
1. Precision/Reproducibility:
A multi-site reproducibility study was performed with the MDx-Chex for BCID2 at four different study sites using three lots of MDx-Chex for BCID2 panel per test site. Testing
K212576 - Page 5 of 15
{5}
consisted of a total of 20 samples being analyzed per lot. Eleven operators participated in the study with 2-4 operator per site. The study was performed on FilmArray instrument 2.0 (used at test sites 1 and 4) and FilmArray Torch systems (used at test sites 1, 2, and 3). A total of 240 external controls were tested. Of the 240 controls tested, 10 positive controls were false negatives, however upon retesting were confirmed as positive controls. As part of the workflow, if an unexpected result is observed for one or more targets within a given run, that sample may be rerun no more than one time to determine if the correct results are obtained for the targets with unexpected results during the first run. Also, of the 10 false negatives, 6 were for not detecting the mecA/C and MREJ antimicrobial resistance gene across three of the four test sites. This resulted in a positive percent agreement of $95.8\%$ $(n = 230 / 240)$ ; see Table 1 below. Also, of the 240 controls tested, 1 negative control was false positive and upon retesting was confirmed as a negative control. This resulted in a negative percent agreement of $99.6\%$ $(n = 239 / 240)$ ; see Table 2 below.
Table 1: Streck Molecular Control (SMC) Reproducibility Summary: Positive Percent Agreement
| Category | Site #1 | | Site #2 | | Site #3 | | Site #4 | | Percent Agreement (all sites combined) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | # Observed Results/# Expected Results1 | Positive Percent Agreement | # Observed Results/# Expected Results1 | Positive Percent Agreement | # Observed Results/# Expected Results1 | Positive Percent Agreement | # Observed Results/# Expected Results1 | Positive Percent Agreement | |
| SMC Level 1 (GN) and Level 2 (GPY), Combined | 59/60* | 98.3% | 57/60* | 95% | 58/60* | 96.7% | 56/60* | 95% | 95.8% (230/240) |
* 10 Positive controls gave initial false negative results; all produced the correct results upon a single retest.
Expected result for the Positive Control is positive. Denominator = total # of results for Level 1-GN and Level 2-GPY controls. GPY = Gram-positive and yeast control, GN = Gram-negative control.
Table 2: Streck Molecular Control (SMC) Reproducibility Summary: Negative Percent Agreement
| Category | Site #1 | | Site #2 | | Site #3 | | Site #4 | | Percent Agreement (all sites combined) |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | # Observed Results/# Expected Results1 | Negative Percent Agreement | #Observed Results/# Expected Results1 | Negative Percent Agreement | # Observed Results/# Expected Results1 | Negative Percent Agreement | # Observed Results/# Expected Results1 | Negative Percent Agreement | |
| SMC Level 1 (GN) and Level 2 (GPY), Combined | 60/60 | 100% | 60/60 | 100% | 60/60 | 100% | 59/60* | 98.3% | 99.6% (239/240) |
* 1 Negative control gave initial false positive result and produced correct result upon a single retest.
Expected result for the Negative Control is negative. Denominator = total # of results for Level 1-GN and Level 2-GPY controls. GPY = Gram-positive and yeast control, GN = Gram-negative control.
The results suggest that there are no significant differences between different users and different study sites on different days. External reproducibility studies for the MDx-Chex for BCID2 are acceptable.
Precision
K212576 - Page 6 of 15
{6}
An internal precision study for the MDx-Chex for BCID2 was conducted over 20 non-consequative days by testing three different control lots. Each lot was tested on two FilmArray Torch instruments. A total of 20 samples were tested per lot and level (Level 1-GN and Level 2-GPY) by four operators. Samples and pouches were prepared according to BCID2 and control instructions. All controls gave correct results except for 6 positive control that gave false negative results, however upon retesting was confirmed as positive controls; This resulted in a positive percent agreement of 95% (n=114/120) and a negative percent agreement of 100% (n= 120/120); see Tables 3 and 4 below.
Table 3: Streck Molecular Control (SMC) Precision Summary: Positive Percent Agreement
| Category | # Observed Results/# Expected Results^{1} | Positive Percent Agreement |
| --- | --- | --- |
| SMC Level 1-GN and Level 2-GPY, Combined | 114/120* | 95% |
* 6 Positive controls gave initial false negative results; all produced the correct results upon a single retest.
Expected result for the Positive Control is positive. Denominator = total # of results for GN (Level 1) and GPY (Level 2) controls. GPY = gram-positive and yeast control, GN = gram negative control.
Table 4: Streck Molecular Control (SMC) Precision Summary: Negative Percent Agreement
| Category | # Observed Results/# Expected Results^{1} | Negative Percent Agreement |
| --- | --- | --- |
| SMC Level 1- GN and Level 2-GPY, Combined | 120/120 | 100% |
Expected result for the Negative Control is negative. Denominator = total # of results for GN (Level 1) and GPY (Level 2) controls. GPY = gram-positive and yeast control, GN = gram negative control.
There appears to be no significant differences in the detection of the control organisms and antimicrobial resistance genes when testing different control lots on different days. Precision studies are acceptable.
Within-run Testing
Within-run precision was demonstrated in a separate study using a single lot of MDx-Chex for BCID2 analyzed on a single day. Six control vials were tested for each control level (Level 1-GN and Level 2-GPY). The samples were analyzed on the FilmArray Torch instrument. The results are shown in Tables 5 and 6 below:
Table 5: Streck Molecular Control (SMC) Within-run Precision Summary: Positive Percent Agreement
| Category | SMC Lot | # Observed Results/# Expected Results^{1} | Positive Percent Agreement |
| --- | --- | --- | --- |
K212576 - Page 7 of 15
{7}
K212576 - Page 8 of 15
| SMC Level 1(GN) and Level 2 (GPY), Combined | 21129 | 12/12 | 100% |
| --- | --- | --- | --- |
¹ Expected result for the Positive Control is positive. Denominator = total # of expected positive result sets for Level 1-GN and Level 2-GPY controls. GN = gram negative control, GPY = gram-positive and yeast control.
Table 6: Streck Molecular Control (SMC) Within-run Precision Summary: Negative Percent Agreement
| Category | SMC Lot | # Observed Results/# Expected Results ¹ | Negative Percent Agreement |
| --- | --- | --- | --- |
| SMC Level 1(GN) and Level 2 (GPY), Combined | 21129 | 12/12 | 100% |
¹ Expected result for the Negative Control is negative. Denominator = total # of expected negative result sets for Level 1-GN and Level 2-GPY controls. GN = gram negative control, GPY = gram-positive and yeast control.
Within-run reproducibility studies for the MDx-Chex for BCID2 are acceptable.
## Lot-to-lot Testing
Lot-to-lot reproducibility was demonstrated by testing three lots of MDx-Chex for BCID2. Samples were prepared and analyzed on the BioFire FilmArray Torch per SMC and BCID2 Instructions for Use. Of the positive controls, one sample gave a false negative result that was confirmed as a positive control upon retesting. Results are shown in Tables 7 and 8 below.
Table 7: Streck Molecular Control (SMC) Lot-to-Lot Precision Summary: Positive Percent Agreement
| Category | SMC Lot | # Observed Results/# Expected Results ¹ | Positive Percent Agreement |
| --- | --- | --- | --- |
| SMC Level 1(GN) and Level 2 (GPY), Combined | 20363 | 11/12* | 91.7% |
| | 20366 | 12/12 | 100% |
| | 21129 | 12/12 | 100% |
* 1 Positive control gave an initial false negative result; it produced the correct result upon a single retest.
¹ Expected result for the Positive Control is positive. Denominator = total # of expected positive result sets for Level 1-GN and Level 2-GPY controls. GN = gram negative control, GPY = gram-positive and yeast control.
Table 8: Streck Molecular Control (SMC) Lot-to-Lot Precision Summary: Negative Percent Agreement
| Category | SMC Lot | # Observed Results/# Expected Results ¹ | Negative Percent Agreement |
| --- | --- | --- | --- |
| SMC Level 1(GN) and Level 2 (GPY), Combined | 20363 | 12/12 | 100% |
| | 20366 | 12/12 | 100% |
{8}
Lot-to-Lot reproducibility studies for the MDx-Chex for BCID2 are acceptable.
2. Linearity:
Not Applicable
3. Analytical Specificity/Interference:
Not Applicable
4. Assay Reportable Range:
Not Applicable
5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
Open Vial Stability
A real-time open-vial stability study is ongoing using three lots of MDx-Chex for BCID2. Samples were stored in a controlled refrigerated (2-8°C) or controlled room temperature (20-25°C). Ten samples per lot per level is being tested at day 0 (baseline= a total of 60 samples per temperature) and a minimum of 61 days for each lot and storage condition. Each lot was tested for different duration (Lot 20263 was tested for 77 days, Lot 20366 was tested for 75 days and Lot 21129 was tested for 63 days) and therefore the shelf-life in Table 9 was denoted as day 61+ for each temperature tested. At baseline, 120 total tubes were tested then divided in half for storage at the two temperatures stated above. All samples are analyzed on the FilmArray Torch instrument. The data presented here are to support a 60 day open-vial stability claim and data to support a longer open-vial stability claim will be submitted at a later date. The data for the 61+ day study showed that at baseline, there were five false negatives and three false positive, that upon retesting were correctly confirmed as positive controls and negative controls, respectively. The data also showed that on day 61+, three samples, stored at 20-25°C, were false negatives and were correctly confirmed as positive controls upon retesting. Therefore, giving an overall PPA and NPA are 96.9% and 99.2%, respectively. The results are shown in Tables 9 and 10 below.
Table 9. Streck Molecular Control (SMC) Open-Vial Stability: Positive Percent Agreement
| Shelf-Life | Storage Temperature | #Observed Results/#Expected Results^{1} | Positive Percent Agreement | PPA ≥ 90% Acceptance |
| --- | --- | --- | --- | --- |
| Day 0* | NA | 115/120** | 95.8% | Pass |
K212576 - Page 9 of 15
{9}
| Day 61+ | 2-8°C | 60/60 | 100% | Pass |
| --- | --- | --- | --- | --- |
| Day 61+ | 20-25°C | 57/60*** | 95% | Pass |
Expected result for the Positive Control is positive. Denominator = total combined # of expected positive results for Level 1-GN and Level 2-GPY controls. GPY = gram-positive and yeast control, GN = gram negative control.
* 120 tubes were completed for Day 0 (baseline). Tubes were divided in half for storage.
** 5 Positive controls gave false negative results on the first test; all reruns produced correct results upon a single retest.
*** 3 positive controls gave false negative results on the first test; all reruns produced correct results upon a single retest.
+ Indicates that each lot was tested for at least 61 days. Lot 20263 (77 days); Lot 20366 (75 days); and Lot 21129 (63 days).
Table 10. Streck Molecular Control (SMC) Open-Vial Stability: Negative Percent Agreement
| Shelf-Life | Storage Temperature | #Observed Results/#Expected Results1 | Negative Percent Agreement | NPA ≥ 90% Acceptance |
| --- | --- | --- | --- | --- |
| Day 0* | NA | 117/120** | 97.5% | Pass |
| Day 61+ | 2-8°C | 60/60 | 100% | Pass |
| Day 61+ | 20-25°C | 60/60 | 100% | Pass |
Expected result for the Negative Control is negative. Denominator = total combined # of expected negative results for Level 1-GN and Level 2-GPY controls. GPY = Gram-positive and yeast control, GN = Gram-negative control.
* 120 tubes were completed for Day 0 (baseline). Tubes were divided in half for storage.
** 3 negative controls gave false positive results on the first test; all reruns produced correct results upon a single retest.
+ Indicates that each lot was tested for at least 61 days. Lot 20263 (77 days); Lot 20366 (75 days); and Lot 21129 (63 days).
The data for the ongoing open-vial stability study support the conclusion that MDx-Chex for BCID2 has an open-vial stability of 60 days when stored at $2 - 25^{\circ}\mathrm{C}$ .
# Closed-vial Stability
A real-time closed-vial stability study is ongoing using three lots of MDx-Chex for BCID2. Samples were stored in a controlled refrigerated $(2 - 8^{\circ}\mathrm{C})$ or controlled room temperature (20- $25^{\circ}\mathrm{C}$ ). Ten samples per lot per level is being tested at day 0 (baseline= a total of 60 samples per temperature) and a minimum of 61 days $(60^{+})$ for each lot and storage condition. At baseline, 120 total tubes were tested then divided in half for storage at the two temperatures stated above. All samples are analyzed on the FilmArray Torch instrument. The data presented here are for at least 61 days and data to support a longer closed-vial stability claim will be submitted at a later date. The same baseline data used for the open-vial stability study above was used here. The closed-vial stability study data showed that on day $61^{+}$ at refrigerated temperature, two samples were false negatives and one sample was false positive, that upon retesting were correctly confirmed as positive controls and negative control, respectively. The data also showed that on day $61^{+}$ at room temperature, one sample was false positive that was correctly confirmed as negative control upon retesting. Therefore, the overall PPA and NPA are $97.5\%$ and $98\%$ , respectively. The results are shown in Tables 11 and 12 below.
Table 11. Streck Molecular Control (SMC) Close-Vial Stability: Positive Percent Agreement
K212576 - Page 10 of 15
{10}
| Shelf-Life | Storage Temperature | #Observed Results/#Expected Results 1 | Positive Percent Agreement | PPA ≥ 90% Acceptance |
| --- | --- | --- | --- | --- |
| Day 0* | NA | 115/120** | 95.8% | Pass |
| Day 61+ | 2-8°C | 58/60*** | 96.7% | Pass |
| Day 61+ | 20-25°C | 60/60 | 100% | Pass |
1 Expected result for the Positive Control is positive. Denominator = total combined # of expected positive results for Level 1-GN and Level 2-GPY controls. GPY = gram-positive and yeast control, GN = gram negative control.
* 120 tubes were completed for Day 0 (baseline). Tubes were divided in half for storage.
** 5 Positive controls gave false negative results on the first test; all reruns produced correct results upon a single retest.
*** 2 Positive controls gave false negative results on the first test; all reruns produced correct results upon a single retest.
+ Indicates that each lot was tested for at least 61 days. Lot 20263 (77 days); Lot 20366 (75 days); and Lot 21129 (63 days).
Table 12. Streck Molecular Control (SMC) Close-Vial Stability: Negative Percent Agreement
| Shelf-Life | Storage Temperature | #Observed Results/#Expected Results 1 | Negative Percent Agreement | NPA ≥ 90% Acceptance |
| --- | --- | --- | --- | --- |
| Day 0* | NA | 117/120** | 97.5% | Pass |
| Day 61+ | 2-8°C | 59/60*** | 98.3% | Pass |
| Day 61+ | 20-25°C | 59/60*** | 98.3% | Pass |
1 Expected result for the Negative Control is negative. Denominator = total combined # of expected negative results for Level 1-GN and Level 2-GPY controls. GPY = Gram-positive and yeast control, GN = Gram-negative control.
* 120 tubes were completed for Day 0 (baseline). Tubes were divided in half for storage.
** 3 negative controls gave false positive results on the first test; all reruns produced correct results upon a single retest.
*** 1 negative control gave a false positive result on the first test; all reruns produced correct results upon a single retest.
*** 1 negative control gave false positive result on the first test; rerun produced correct result upon a single retest.
+ Indicates that each lot was tested for at least 61 days. Lot 20263 (77 days); Lot 20366 (75 days); and Lot 21129 (63 days).
The data for the ongoing closed-vial stability study support the conclusion that MDx-Chex for BCID2 has a closed-vial stability of 60 days when stored at 2-25°C.
## Real-time Stability Testing
Real-time stability studies are ongoing to support product claims for the MDx-Chex for BCID2. Real-time stability study protocols and acceptance criteria were reviewed and found to be acceptable.
## Shipping Stability
A shipping stability study was performed to validate the stability of MDx-Chex for BCID2 during shipment. Using a single lot, two sets of samples were simulated into a summer or winter months profile over five days. Temperature stress conditions for Summer followed a
K212576 - Page 11 of 15
{11}
120-hour exposure period with the following temperature profile: 5h at 22°C, 14h at 26°C, 5.5h at 22°C, 5.5h at 25°C, 10h at 22°C, 2h at 40°C, 15h at 29°C, 5h at 40°C, 17.5h at 26°C, 11.5h at 29°C, 10h at 22°C, 7.5h at 30°C, and 11.5h at 24°C. Temperature stress conditions for Winter followed a 120-hour exposure with the following temperature profile: 5h at 22°C, 14h at 15°C, 5.5h at 22°C, 5.5h at 17°C, 10h at 22°C, 2h at -10°C, 15h at 10°C, 5h at -10°C, 17.5h at 15°C, 11.5h at 10°C, 10h at 22°C, 7.5h at 8°C, and 11.5h at 18°C. After temperature stress, samples were then incubated at 2-8°C for a week prior to being tested on the FilmArray Torch instrument. Ten samples were analyzed for each control (Level 1 and Level 2) for a total of 20 samples analyzed. Data showed that analysis using the summer profile resulted in one false negative and one false positive that upon retesting was correctly confirmed as a positive control and a negative control, respectively. Therefore, the overall PPA and NPA were 97.5% individually. Results are presented in Tables 13 and 14 below.
Table 13: Summary of Shipping Study for the Streck Molecular Control MDx-Chex: Positive Agreement
| Category | # Expected results / # Tested^{1} | Positive Percent Agreement |
| --- | --- | --- |
| Summer | 19/20* | 95% |
| Winter | 20/20 | 100% |
1 Expected result for the Positive Control is positive.
* 1 Positive control gave initial false negative result; it produced the correct result upon a single retest.
Table 14: Summary of Shipping Study for the Streck Molecular Control MDx-Chex: Negative Agreement
| Category | # Expected results / # Tested^{1} | Negative Percent Agreement |
| --- | --- | --- |
| Summer | 19/20* | 95% |
| Winter | 20/20 | 100% |
1 Expected result for the Negative Control is negative.
* 1 Negative control gave initial false positive result; it produced correct result upon a single retest.
The data support a claim that MDx-Chex for BCID2 remains functional after exposure to extreme temperature that may occur during shipping.
6. Detection Limit:
Not Applicable
7. Assay Cut-Off:
Not Applicable
K212576 - Page 12 of 15
{12}
K212576 - Page 13 of 15
# B Comparison Studies:
1. Method Comparison with Predicate Device:
Not Applicable
2. Matrix Comparison:
The matrix of MDx-Chex for BCID2 is made of stabilized blood components (erythrocytes and leukocytes) in a simulated blood culture media. Since the matrix is not identical to that of the routine BCID2 assay sample, blood culture media, a test was performed to investigate the effect of the matrix on the assay.
To confirm that the matrix has no effect on the assay or control, inactivated Streptococcus pneumoniae was spiked into the MDx-Chex for BCID2 matrix as well as into BD BACTEC Plus Aerobic/F culture vial with negative whole blood to simulate a clinical sample. These samples were then tested in triplicate using the BCID2 Panel. Three replicates of each simulated matrix with no spike-in organism were also tested to serve as negative controls. Samples were analyzed on FilmArray Torch instrument. The results showed PPA and NPA of 100% and results are shown in Tables 15 and 16 below.
Table 15: Effect of Streck Molecular Control MDx-Chex and Clinical Samples Spiked with S. pneumonia, Tested on BCID-2 panel
| Matrix type | # Expected results / # tested 1 | Percent Agreement |
| --- | --- | --- |
| MDx-Chex, Positive Matrix | 3/3 | 100% |
| Clinical, Positive Matrix | 3/3 | 100% |
1 Expected result for the Positive Matrix is positive.
Table 16: Effect of Negative Streck Molecular Control MDx-Chex and Negative Clinical Samples, Tested on BCID-2 panel
| Matrix type | # Expected results / #tested 1 | Percent Agreement |
| --- | --- | --- |
| MDx-Chex, Negative Matrix | 3/3 | 100% |
| Clinical, Negative Matrix | 3/3 | 100% |
1 Expected result for Negative Matrix is negative.
The results show that samples prepared with the MDx-Chex for BCID2 matrix showed no inhibition and/or false negative results when used with the FilmArray BCID2 panel. The data
{13}
also support the conclusion that the SMC MDx-Chex performs identically to a clinical BCID2 panel sample, positive blood culture sample.
# C Clinical Studies:
1. Clinical Sensitivity:
Not Applicable
2. Clinical Specificity:
Not Applicable
3. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable):
Not Applicable
# D Clinical Cut-Off:
Not Applicable
# E Expected Values/Reference Range:
MDx-Chex for BCID2 is a qualitative control and the expected results are listed in Table 17 below.
Table 17 - Pathogens and antimicrobial resistance genes detected by MDx-Chex for BCID2 Control
| Gram-Negative Bacteria | | |
| --- | --- | --- |
| Pathogen | MDx-Chex for BCID2 Control 1 - GN | MDx-Chex for BCID2 Control 2 - GPY |
| Acinetobacter colcoaceticus-baumannii complex | Detected | Not detected |
| Bacteroides fragilis | Detected | Not detected |
| Enterobacter cloacae complex | Detected | Not detected |
| Escherichia coli | Detected | Not detected |
| Klebsiella aerogenes | Detected | Not detected |
| Klebsiella oxytoca | Detected | Not detected |
| Klebsiella pneumoniae group | Detected | Not detected |
| Proteus spp. | Detected | Not detected |
| Salmonella spp. | Detected | Not detected |
| Serratia marcescens | Detected | Not detected |
| Haemophilus influenzae | Detected | Not detected |
| Neisseria meningitides | Detected | Not detected |
| Pseudomonas aeruginosa | Detected | Not detected |
| Stenotrophomonas maltophilia | Detected | Not detected |
| Gram-Positive Bacteria | | |
| Pathogen | MDx-Chex for BCID2 Control 1 - GN | MDx-Chex for BCID2 Control 2 - GPY |
| Enterococcus faecalis | Not detected | Detected |
| Enterococcus faecium | Not detected | Detected |
K212576 - Page 14 of 15
{14}
| Yeast | | |
| --- | --- | --- |
| Pathogen | MDx-Chex for BCID2 Control 1 - GN | MDx-Chex for BCID2 Control 2 - GPY |
| Candida albicans | Not detected | Detected |
| Candida auris | Not detected | Detected |
| Candida glabrata | Not detected | Detected |
| Candida krusei | Not detected | Detected |
| Candida parapsilosis | Not detected | Detected |
| Candida tropicalis | Not detected | Detected |
| Cryptococcus neoformans/gatti | Not detected | Detected |
| Antimicrobial Resistance Genes | | |
| --- | --- | --- |
| Gene | MDx-Chex for BCID2 Control 1 - GN | MDx-Chex for BCID2 Control 2 - GPY |
| mecA/C and MREJ | Not detected | Detected |
| vanA/B | Not detected | Detected |
| KPC | Detected | Not detected |
| CTX-M | Detected | Not detected |
| IMP | Detected | Not detected |
| NDM | Detected | Not detected |
| OXA-48-like | Detected | Not detected |
| VIM | Detected | Not detected |
| mcr-1 | Detected | Not detected |
# VIII Proposed Labeling:
The labeling supports the finding of substantial equivalence for this device.
# IX Conclusion:
The submitted information in this premarket notification is complete and does support a substantial equivalence decision.
K212576 - Page 15 of 15
Predicate graph will load when search results are available.
Embedding visualization will load when search results are available.
PDF viewer will load when search results are available.
Loading panels...
Select an item from Submissions
Click any panel, subpart, regulation, product code, or device to see details here.
Section Matches
Results will appear here.
Product Code Matches
Results will appear here.
Special Control Matches
Results will appear here.
Loading collections...
Loading
My Alerts
You will receive email notifications based on the filters and frequency you set for each alert.
Sort by:
Create Alert
Search Filters
Agent Token
Create a read-only bearer token for Claude, ChatGPT, or other agents that can call HTTP APIs.
Copy this now. It will not be shown again.
Connected apps
Apps you authorized through browser sign-in. Disconnecting revokes their access immediately.
Learn the FDA Browser
Two short videos show you everything — or skip straight to the written tutorial if you'd rather read. You can reopen this any time from the Tutorial button in the top bar.
Part 1 — Search, results, and everyday workflows 16 min
Part 2 — Embeddings: the galaxy map 3 min
1. Search: exact and fuzzy
Type a phrase like "coronary artery calcification" into the search box. You get two kinds of results. Exact results match the literal phrase — prefix searches work ("coronary artery calcificati") but suffix searches do not. Fuzzy results match on the meaning and intent of your phrase rather than the exact words, and are sorted by relevance score. Hover over the Exact or Fuzzy badge on any row to see exactly why it matched.
Use the checkboxes above the results to narrow: SaMD keeps only software-only devices, AI / ML keeps only devices with AI.
Exact vs. fuzzy search: what's the difference?
Exact matches on the literal phrase (prefix search works, suffix does not). Fuzzy matches on the meaning and intent of the phrase rather than the exact words. Hover over the badge on any row to see why it matched.
You search "coronary artery calcification" and want only software devices with AI. What two filters do you apply?
Narrow by SaMD (software-only devices), then narrow by AI/ML (devices with AI).
2. The results table
Scroll right in the results table. The intended use is extracted for you — no need to open the PDF. The device story gives a high-level snapshot of what the device does and how it's used. The AI Performance sub-table shows each output name, acceptance criteria, observed values, and development/test dataset descriptions — the same format Innolitics uses for regulatory strategy outputs, and the fastest high-level fingerprint of an AI device. It is AI-generated but has been very reliable in practice.
Where do you find a device's intended use without opening the PDF?
Scroll right in the search results table. The intended use column is extracted for you; no need to dig into the 510(k) summary PDF.
What does the AI Performance sub-table show, and why is it useful?
Output name, acceptance criteria, observed values, development dataset description, and test dataset description. It's the same format we use for regulatory strategy output and Fast 510(k) input, and the fastest high-level fingerprint of an AI device. AI-generated but reliable in practice.
3. Judging fuzzy relevance
Fuzzy results trail off in relevance as you scroll. Use three signals to decide how far down to go: the fuzzy badge explanations, the intended use column, and whether your target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, you're past the relevant zone. A top hit with a low score (~0.4) and a stretched explanation is a hint the closest predicates are far away — the project may be headed for De Novo. Note the fuzzy search is a pattern match: it doesn't handle negation ("not") well, and hardware devices can appear — filter by SaMD/AI ML to cut them.
How do you judge how far down fuzzy search results to go?
Use the relevancy signals: the fuzzy badge explanations, the intended use column, and whether the target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, results are trailing off in relevancy.
4. Device detail page: chat and citations
Click a device name to open its detail page: device facts on the left, a chat window on the right. Ask something like "Describe the training data". The answer carries little citation bubbles — click one to jump to the highlighted passage in the source PDF, so you can verify every AI answer against the document. There's also a Download PDF button for sharing.
How do you verify an AI chat answer on the device detail page?
Click the citation bubbles to jump to the relevant highlight in the source document.
Reading rule for every project: how many summaries do you read in full?
At least the three most relevant 510(k) or De Novo summaries, in full. After that, use targeted chat questions to confirm your memory quickly. The tool supports this professional habit — it doesn't replace it.
5. Side-by-side comparison
Select multiple rows in the results table (aim for under ~10), then open the PDF Viewer tab. Ask one question — it goes to all selected devices in parallel, each with citations. This is the fastest way to compare and contrast devices: training data, PCCP scope, how they handled adding new scanners, and so on.
What does the side-by-side PDF viewer mode do?
Select multiple devices, open the PDF viewer tab, and ask one question (e.g., "Describe the training data"). It queries all selected devices simultaneously with citations, so you can compare and contrast quickly.
6. Collections
With rows selected, go to the Collections tab and create a labeled collection (e.g., "Cobb Angle Project"). Reload that selection any time — before a client call, pull up the collection and ask questions across all of its devices at once.
How do you save a set of selected devices for later use?
Select the rows, go to the Collections tab, and create a labeled collection (e.g., "Cobb Angle Project"). You can reload the selection anytime and carry it into the PDF viewer and other tabs that support selections.
7. Product codes and the regulations tree
Click a product code in the results to jump to it in the regulations tree — identification text, sibling product codes, and devices you can open in a PDF viewer on the right. Click a regulation number to see its identification, special controls, and related product codes. You can also search by product code or regulation number at the top of the tree. Always read the special controls if any exist for your device — it broadens your search and sharpens pre-kickoff research.
What can you do from the regulations tree view?
Browse product codes and regulation numbers, read the identification text and special controls, browse sibling product codes, open device PDFs on the right, and search by product code or regulation number at the top of the tree.
8. Chart view
Click Show Chart and segment by regulation number (or product code) to see which regulations dominate your result set. Clicking a regulation takes you into the regulations tree. Great for spotting that most matches are, say, hardware laparoscopic devices — a cue to go back and filter.
How do you see which regulations dominate a search result set?
Click "Show Chart" and segment by Regulation Number. Clicking a regulation takes you to the regulations tree.
9. The predicate graph
Open the Predicates tab for a family-tree view of predicate relationships. Click a node to trace its parents and children; selections from search carry over pre-selected. Commonly predicated devices are worth reading — a lot of people predicated them for a reason. The visual lineage is also handy on client calls, e.g. to show how a predicate family evolved and justify why your predicate still holds.
In the predicate graph, why are commonly predicated devices worth reading?
A lot of people predicated them for a reason. Clicking a node traces parents and children, and selections from search carry over pre-selected.
10. Embeddings: the galaxy map
The Embeddings tab plots every matching document in a 2-D "galaxy map" where semantically similar devices cluster together. Hover or click clusters to explore, and let AI label the clusters for you. Embeddings beat product codes for grouping: two devices can carry different product codes (LLZ vs. QIH) yet do the same thing — the embedding captures the meaning of the intended use and device story. This is also exactly how retrieval-augmented generation (RAG) works under the hood, and it makes a great visual on client calls.
Try it yourself
Head to the search page and work through a few of these AI/ML fuzzy searches to build intuition: perivascular fat on CT · aortic valve calcification opportunistic screening on noncontrast CT · breast cancer prediction on digital pathology slides · autism detection · gestational age prediction · a hearing aid that can also detect a pulse · foundation model based analysis of ECG · large language models · penetration test. Watch how the relevance scores, intended use, and AI Performance tables tell you when results stop being meaningful.