K231616 · Zeus Scientific · KTL · Aug 31, 2023 · Immunology
Device Facts
Record ID
K231616
Device Name
ZEUS IFA(TM) nDNA Test System, ZEUS dIFine
Applicant
Zeus Scientific
Product Code
KTL · Immunology
Decision Date
Aug 31, 2023
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.5100
Device Class
Class 2
Indications for Use
The ZEUS IFA™ nDNA Test System is an indirect immunofluorescence assay utilizing Crithidia luciliae for the qualitative and semi-quantitative determination of anti-native DNA (nDNA) IgG antibodies to DNA in human serum by manual fluorescence microscopy or with ZEUS dIFine®. The presence of nDNA antibodies in conjunction with other serological and clinical findings can be used to aid in the diagnosis of systemic lupus erythematosus (SLE).
Device Story
The ZEUS IFA nDNA Test System is an indirect immunofluorescence assay (IFA) using Crithidia luciliae substrate to detect anti-native DNA (nDNA) IgG antibodies in human serum. The device is used in clinical laboratories by trained personnel. Samples are incubated with the substrate; if nDNA antibodies are present, they bind to the kinetoplast of the Crithidia luciliae. Bound antibodies are detected using a fluorescently labeled anti-human IgG conjugate. Results are visualized via manual fluorescence microscopy or processed using the ZEUS dIFine® automated system. The output provides qualitative or semi-quantitative antibody levels, which clinicians use alongside other serological and clinical data to support a diagnosis of systemic lupus erythematosus (SLE).
Clinical Evidence
Clinical performance was evaluated using 660 clinically characterized samples (300 SLE, 360 non-SLE controls). Sensitivity and specificity were compared across three methods. Sensitivity for SLE ranged from 22.33% to 27.00% across methods and sites. Specificity ranged from 97.50% to 99.72%. Method comparison studies (N=3,960 comparisons) showed high agreement between manual and automated methods, with overall agreement between Method A and Method C (UNC as positive) at 98.31% (95% CI: 97.86%–98.66%).
Technological Characteristics
Indirect immunofluorescence assay using C. luciliae substrate slides. Reagents include FITC-labeled goat anti-human IgG conjugate, positive/negative human serum controls, and PBS. ZEUS dIFine system uses a 20X objective to capture 12 fields of view (approx. 610 µm x 510 µm) per well. Software performs automated image analysis and fluorescence intensity measurement. System is intended for clinical laboratory use.
Indications for Use
Indicated for qualitative and semi-quantitative detection of anti-native DNA (nDNA) IgG antibodies in human serum to aid in the diagnosis of systemic lupus erythematosus (SLE). For use by trained operators in clinical laboratory settings. Prescription use only.
Regulatory Classification
Identification
An antinuclear antibody immunological test system is a device that consists of the reagents used to measure by immunochemical techniques the autoimmune antibodies in serum, other body fluids, and tissues that react with cellular nuclear constituents (molecules present in the nucleus of a cell, such as ribonucleic acid, deoxyribonucleic acid, or nuclear proteins). The measurements aid in the diagnosis of systemic lupus erythematosus (a multisystem autoimmune disease in which antibodies attack the victim's own tissues), hepatitis (a liver disease), rheumatoid arthritis, Sjögren's syndrome (arthritis with inflammation of the eye, eyelid, and salivary glands), and systemic sclerosis (chronic hardening and shrinking of many body tissues).
Submission Summary (Full Text)
{0}
FDA
U.S. FOOD & DRUG
ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY ONLY
## I Background Information:
A 510(k) Number
K231616
B Applicant
ZEUS Scientific
C Proprietary and Established Names
ZEUS IFA nDNA Test System
ZEUS dIFine
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| KTL | Class II | 21 CFR 866.5100 - Antinuclear Antibody Immunological Test System | IM - Immunology |
| PIV | Class II | 21 CFR 866.4750 - Automated indirect immunofluorescence microscope and software-assisted system | IM - Immunology |
## II Submission/Device Overview:
A Purpose for Submission:
Migration of previously cleared assay to a previously cleared instrument
B Measurand:
Anti-double stranded DNA (dsDNA) IgG autoantibodies
C Type of Test:
Qualitative and/or semi-quantitative indirect immunofluorescence (IFA); manual or semi-automated
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
K231616 - Page 2 of 17
## III Intended Use/Indications for Use:
### A Intended Use(s):
See Indications for Use below.
### B Indication(s) for Use:
The ZEUS IFA nDNA Test System is an indirect immunofluorescence assay utilizing Crithidia luciliae for the qualitative and semi-quantitative determination of anti-native DNA (nDNA) IgG antibodies to DNA in human serum by manual fluorescence microscopy or with ZEUS dIFine. The presence of nDNA antibodies in conjunction with other serological and clinical findings can be used to aid in the diagnosis of systemic lupus erythematosus (SLE).
### C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
- This device is only for use with reagents that are indicated for use with the device.
- The device is for use by a trained operator in a clinical laboratory setting.
- All software-aided results must be confirmed by a trained operator.
### D Special Instrument Requirements:
For use only with Zeus dIFine
## IV Device/System Characteristics:
### A Device Description:
ZEUS IFA nDNA Test System is an indirect immunofluorescence assay for the qualitative detection and semi-quantitative determination of anti-native DNA (nDNA) IgG antibodies in human serum by manual fluorescence microscopy or with ZEUS dIFine.
#### ZEUS IFA nDNA Test System Assay kit components
1. C. luciliae Substrate Slides: Twenty, 10-well Slides with blotter.
2. Conjugate: Goat anti-human IgG labeled with FITC. Contains phosphate buffer with BSA and counterstain. One bottle (white-capped amber bottle), 12 mL. Ready to use.
3. Positive Control (Human Serum): Will produce positive apple-green staining of the kinetoplast in the C. luciliae organisms. One vial (red-capped), 0.5 mL. Ready to use.
4. Negative Control (Human Serum): Will produce no detectable nDNA staining. One vial (green-capped), 0.5 mL. Ready to use.
5. SAVe Diluent: Four bottles (green-capped), 30 mL phosphate-buffered-saline. Ready to use.
6. Phosphate-buffered-saline (PBS): pH 7.2 ± 0.2. Ten packets, each sufficient to prepare one liter PBS in distilled or deionized water.
7. Mounting Media (Buffered Glycerol): One bottle (clear-capped clear bottle), 12 mL. Ready to use.
{2}
B Principle of Operation:
The ZEUS IFA nDNA Test System is an indirect fluorescent antibody assay for the qualitative and semi-quantitative determination of anti-nDNA IgG antibody in human sera by manual fluorescence microscopy or by ZEUS dIFine. The reaction occurs in two steps:
1. Step one: If nDNA antibodies are present in a sample, a reaction between nDNA antibodies and the kinetoplast of the *C. luciliae* substrate takes place in the first step.
2. Step two: Goat anti-human IgG labeled with fluorescein isothiocyanate (FITC) is added to the substrate. If the patient’s sera contain anti-nDNA IgG antibody, a positive apple-green, fluorescent antigen-antibody reaction will be observed when the slides are examined with the fluorescence microscope. Smooth or peripheral staining of the kinetoplast with a bright fluorescence located near the flagellar region of the *C. luciliae* is considered a positive reaction.
Manual Interpretation: The interpretation of the results depends on the pattern observed as well as the titer of the autoantibody present in the specimen. A positive reaction is the presence of any pattern of nuclear apple-green staining observed at a 1:10 dilution based on a 1+ to 4+ scale of staining intensity where 1+ is considered a weak reaction and 4+ a strong reaction. The sponsor recommends sera positive at 1:10 should be titered to endpoint dilution by making 1:20, 1:40, 1:80, etc. serial dilutions. The endpoint titer is the highest dilution that produces a 1+ positive reaction.
Zeus dIFine Interpretation: When slides are analyzed by Zeus dIFine, digital images of representative fields of view of the well are captured. The default scanning area is composed of 12 fields with each field approximately 610 µm x 510 µm using the 20X objective. All images are taken through a fluorescein isothiocyanate (FITC) filter. The ZEUS dIFine System reads the slides and measures the FITC light intensity of the *Crithidia* that are included in the region. Then, the ZEUS dIFine System reports the measured average nuclear fluorescence intensity as an index percentage and recommends a qualitative result. To facilitate the interpretation, the ZEUS dIFine System presents acquired digital images of the slide and the following information for human analysis: Negative, Positive or Uncertain (only for Zeus IFA nDNA Test System on ZEUS dIFine with automated reading (software interpretation) of the slides) classification
Trained operators then review the images taken by ZEUS dIFine System. During the review process, further options include:
- Navigation of digitized well using virtual microscope tools (zooming/browsing images at different magnifications).
- Enlargement of images to examine detail.
- Image Atlas to assist with identification of patterns.
The trained operator can confirm the results by clicking the “Validate” button on the screen and accepting the classification (negative/positive and pattern) suggested by the ZEUS dIFine System, or they can revise the suggested ZEUS dIFine classification (negative to positive and vice versa or from Uncertain to either positive or negative), add comments (if any) and eventually click the “Validate” button on the screen.
K231616 - Page 3 of 17
{3}
V Substantial Equivalence Information:
A Predicate Device Name(s):
Test System For nDNA Antibody Determination
B Predicate 510(k) Number(s):
K780178
C Comparison with Predicate(s):
| Device & Predicate Device(s): | K231616 | K780178 |
| --- | --- | --- |
| Device Trade Name | Zeus IFA nDNA Test System, Zeus dIFine | Zeus IFA nDNA Test System |
| General Device Characteristic Similarities | | |
| Intended Use/Indications for Use | The ZEUS IFA nDNA Test System is an indirect immunofluorescence assay utilizing Crithidia luciliae for the qualitative and semi quantitative determination of anti-native DNA (nDNA) IgG antibodies to DNA in human serum by manual fluorescence microscopy or with ZEUS dIFine. The presence of nDNA antibodies in conjunction with other serological and clinical findings can be used to aid in the diagnosis of systemic lupus erythematosus (SLE). | This is an indirect fluorescent antibody test for the semi-quantitative detection of IgG anti-nDNA antibodies in human serum. This test system is to be used as an aid in the diagnosis of systemic lupus erythematosus. |
| Methodology | Indirect immunofluorescence assay | Same |
| Sample Matrix | Serum | Same |
| Fluorescence Marker | FITC | Same |
| Control | One positive control and one negative control | Same |
| Screening Dilution | 1:10 | Same |
| Antigen | Crithidia luciliae | Same |
| Results | Qualitative, semi-quantitative | Same |
| Storage Conditions | Unopened test system, mounting media, Conjugate Zorba-NS, Slides, Positive control, and Negative control must be stored at 2-8 °C | Same |
| General Device Characteristic Differences | | |
| Interpretation of results | Manual fluorescence microscopy or Zeus dIFine automated microscopy with trained operator verification | Manual fluorescence microscopy |
K231616 - Page 4 of 17
{4}
VI Standards/Guidance Documents Referenced:
The following Clinical and Laboratory Standards Institute (CLSI) guidelines were used:
- CLSI EP05-A3, Evaluation of Precision Performance of Quantitative Measurement Methods, Approved Guideline – Third Edition
- CLSI EP06-Ed2, Evaluation of the Linearity of Quantitative Measurement Procedures – Second Edition
- CLSI EP07-A3, Interference Testing in Clinical Chemistry – Third Edition
- CLSI EP17-A2, Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures – Second Edition
- CLSI EP28-A3c, Defining, Establishing and Verifying Reference Intervals in the Clinical Laboratory
VII Performance Characteristics (if/when applicable):
All analytical and clinical studies were evaluated by comparing the three possible reading methods (A, B, and C) described in the table below; these methods are consistent throughout this document. Method A (i.e., manual imaging and manual reading of the slides with a traditional fluorescence microscope) is considered the reference method (predicate) to which all results are compared. All results generated by the Zeus IFA nDNA Test System must be confirmed by a trained operator.
Table 1. Interpretation Methods
| Method | Processing | Imaging | Reading/Evaluation of Slides |
| --- | --- | --- | --- |
| A (predicate) | Manual | Manual | Manual (read of microscope field) |
| B | Automated | Automated | Manual (read of digital image) |
| C | Automated | Automated | Automated (software interpretation) |
A Analytical Performance:
1. Precision/Reproducibility:
a. Repeatability - 20-Day Within-Laboratory Precision Study
Two low positive serum samples (~1:10/1:20 endpoint), two medium positive serum sample (~1:40 to 1:80 endpoint), two high positive serum samples (>1:160 endpoint) and two negative specimens were assayed in triplicate, on 20 different days producing 60 results per sample at one site. The slides were interpreted by all three methods for qualitative results. All slides were interpreted via Methods A and B independently by two technicians, as well as Method C using a single instrument.
K231616 - Page 5 of 17
{5}
Within-Method Qualitative Result Agreement:
There was 100% within-method qualitative result agreement with the expected result for all eight samples when interpreted via Methods A and B, for both technicians. For Method C, the medium positive sample#1, high positive sample#2, and both negative samples yielded 100% within-method qualitative result agreement. The low positive sample#1, low positive sample#2, medium positive sample#2, and high positive sample#1 yielded within-method qualitative result agreement values of 96.7%, 98.3%, 98.3%, and 95.0%, respectively. The results are summarized in the following tables:
Table 2: Within-Method Qualitative Result Agreement
| Sample | Agreement (95% CI) | | |
| --- | --- | --- | --- |
| | Method A | Method B | Method C |
| Low Positive-1 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 96.7%
(88.6 - 99.1%) |
| Low Positive-2 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 98.3%
(91.1 - 99.7%) |
| Medium Positive-1 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) |
| Medium Positive-2 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 98.3%
(91.1 - 99.7%) |
| High Positive-1 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 95.0%
(86.3 - 98.3%) |
| High Positive-2 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) |
| Negative-1 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) |
| Negative-2 | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) | 100%
(88.7 - 100%) |
Between-Method Agreement:
There was 100% between-method qualitative result agreement for all eight samples when interpreted via Method A versus Method B, for both technicians. When Method A and Method B were compared to Method C, the medium positive sample#1, high positive sample#2, and both negative samples yielded 100% between-method qualitative result agreement. The low positive sample#1, low positive sample#2, medium positive sample#2, and high positive sample#1 yielded between-method qualitative result agreement values of 96.7%, 98.3%, 98.3%, and 95.0%, respectively. The results are summarized in the following tables:
K231616 - Page 6 of 17
{6}
Table 3: Between-method Qualitative Agreement
| Sample | Agreement (95% CI) | | | |
| --- | --- | --- | --- | --- |
| | | Method A vs Method B | Method A vs Method C | Method B vs Method C |
| Low Positive-1 | | 100% (88.7 - 100%) | 96.7% (88.6 - 99.1%) | 96.7% (88.6 - 99.1%) |
| Low Positive-2 | | 100% (88.7 - 100%) | 98.3% (91.1 - 99.7%) | 98.3% (91.1 - 99.7%) |
| Medium Positive-1 | | 100% (88.7 - 100%) | 100% (88.7 - 100%) | 100% (88.7 - 100%) |
| Medium Positive-2 | | 100% (88.7 - 100%) | 98.3% (91.1 - 99.7%) | 98.3% (91.1 - 99.7%) |
| High Positive-1 | | 100% (88.7 - 100%) | 95.0% (86.3 - 98.3%) | 95.0% (86.3 - 98.3%) |
| High Positive-2 | | 100% (88.7 - 100%) | 100% (88.7 - 100%) | 100% (88.7 - 100%) |
| Negative-1 | | 100% (88.7 - 100%) | 100% (88.7 - 100%) | 100% (88.7 - 100%) |
| Negative-2 | | 100% (88.7 - 100%) | 100% (88.7 - 100%) | 100% (88.7 - 100%) |
# b. Reproducibility - 5-day Site-to-Site Reproducibility Study
Two negative serum samples, two low positive serum samples ( $\sim 1:10-1:20$ endpoint), two medium positive serum samples ( $\sim 1:40-1:80$ endpoint), and two strong positive serum samples ( $>1:320$ endpoint) were assayed at a 1:10 screening dilution in triplicate, twice per day, on five different days, at three different laboratories. Qualitative results were interpreted by two technicians at each laboratory for Methods A and B, and by a single dIFine instrument at each laboratory for Method C. The results are summarized in the following tables.
Table 4: Qualitative Agreement for Method A between Technicians Across Three Sites:
| Method A | Site 1 | | Site 2 | | Site 3 | | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | Technician 1 | Technician 2 | Technician 1 | Technician 2 | Technician 1 | Technician 2 |
| Site 1 | Technician 1 | | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) |
| | Technician 2 | | | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) |
| Site 2 | Technician 1 | | | | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) |
| | Technician 2 | | | | | 100% (98.42-100.00) | 100% (98.42-100.00) |
| Site 3 | Technician 1 | | | | | | 100% (98.42-100.00) |
| | Technician 2 | | | | | | |
K231616 - Page 7 of 17
{7}
Table 5: Qualitative Agreement for Method B between Technicians Across Three Sites:
| Method B | Site 1 | | Site 2 | | Site 3 | | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | Technician 1 | Technician 2 | Technician 1 | Technician 2 | Technician 1 | Technician 2 |
| Site 1 | Technician 1 | | 100% (98.42-100.00) | 100% (98.42-100.00) | 100% (98.42-100.00) | 99.17% (97.01-99.77) | 99.17% (97.01-99.77) |
| | Technician 2 | | | 100% (98.42-100.00) | 100% (98.42-100.00) | 99.17% (97.01-99.77) | 99.17% (97.01-99.77) |
| Site 2 | Technician 1 | | | | 100% (98.42-100.00) | 99.17% (97.01-99.77) | 99.17% (97.01-99.77) |
| | Technician 2 | | | | | 99.17% (97.01-99.77) | 99.17% (97.01-99.77) |
| Site 3 | Technician 1 | | | | | | 100% (98.42-100.00) |
| | Technician 2 | | | | | | |
Table 6: Qualitative Agreement for Method C across Three Sites:
| Method C | Site 1 | Site 2 | Site 3 |
| --- | --- | --- | --- |
| Site 1 | | 94.58% (90.95 - 96.81) | 95.83% (92.50 - 97.72) |
| Site 2 | | | 96.25% (93.03 - 98.01) |
| Site 3 | | | |
c. Lot-to-Lot Reproducibility
Nine negative serum samples, two low positive serum samples (~1:10-1:20 endpoint), two medium positive serum samples (~1:40-1:80 endpoint), and two strong positive serum samples (> 1:320 endpoint) were assayed at a 1:10 screening dilution using three different reagent lots. For the six positive samples, additional serial dilutions ranging from 1:20 through 1:5120, were also assayed and interpreted by all three methods, towards determining an endpoint titer. There was 100% agreement in the qualitative results at the screening dilution for all 15 specimens across all three kit lots, for interpretation methods A and B. For lot 3, one UNC result was obtained via interpretation method C at the 1:10 screening dilution, for one of the low positive samples. All six positive specimens resulted in the same endpoint titers ± one dilution regardless of reagent kit lot or method interpretation.
2. Linearity:
Two low positive serum samples (~1:10-1:20 endpoint), two medium positive serum samples (~1:40-1:80 endpoint), and two strong positive serum samples (>1:320 endpoint) were assayed at a 1:10 screening dilution, as well as at serial dilutions ranging from 1:20 through 1:5120, then interpreted by all three methods. Consistent positivity was found throughout dilution to endpoint and a consistent titer endpoint was found between methods (within ±1 titer). Individual results per dilution were read reported using a fluorescence intensity scale
K231616 - Page 8 of 17
{8}
from "0" (negative) to "4" (high positive). The endpoints for each sample and each method are presented below:
Table 7: Endpoint Titers across Methods
| Sample | Method A | Method B | Method C |
| --- | --- | --- | --- |
| Low Positive-1 | 1:10 | 1:20 | 1:20 |
| Low Positive-2 | 1:20 | 1:40 | 1:40 |
| Medium Positive-1 | 1:40 | 1:80 | 1:80 |
| Medium Positive-2 | 1:40 | 1:80 | 1:80 |
| High Positive-1 | 1:640 | 1:640 | 1:640 |
| High Positive-2 | 1:640 | 1:640 | 1:640 |
Table 8: The sample dilutions and associated intensity grade results for Method A
| Sample | 1:10 | 1:20 | 1:40 | 1:80 | 1:160 | 1:320 | 1:640 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Low Positive-1 | 1 | 0 | 0 | 0 | 0 | NT | NT |
| Low Positive-2 | 2 | 1 | 0 | 0 | 0 | NT | NT |
| Medium Positive-1 | 2 | 1 | 1 | 0 | 0 | 0 | 0 |
| Medium Positive-2 | 2 | 1 | 1 | 0 | 0 | 0 | 0 |
| High Positive-1 | 4 | 3 | 3 | 2 | 2 | 1 | 1 |
| High Positive-2 | 4 | 3 | 3 | 2 | 2 | 1 | 1 |
Table 9: The sample dilutions and associated intensity grade results for Method B
| Sample | 1:10 | 1:20 | 1:40 | 1:80 | 1:160 | 1:320 | 1:640 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Low Positive-1 | 1 | 1 | 0 | 0 | 0 | NT | NT |
| Low Positive-2 | 2 | 1 | 1 | 0 | 0 | NT | NT |
| Medium Positive-1 | 2 | 2 | 1 | 1 | 0 | 0 | 0 |
| Medium Positive-2 | 2 | 2 | 1 | 1 | 0 | 0 | 0 |
| High Positive-1 | 4 | 3 | 3 | 2 | 2 | 1 | 1 |
| High Positive-2 | 4 | 3 | 3 | 2 | 2 | 1 | 1 |
Method C cannot provide an intensity grade, only a positive, negative, or uncertain call.
# 3. Analytical Specificity/Interference:
# a. Interference studies
Two negative serum samples, two low positive serum samples ( $\sim 1:10-1:20$ endpoint), two medium positive serum samples ( $\sim 1:40-1:80$ endpoint), and two strong positive serum samples ( $>1:320$ endpoint) were spiked with two different concentrations (low spike and high spike) of the 19 different interferents outlined in the table below. All specimens were assayed in triplicate by the ZEUS IFA nDNA Test System and interpreted by all three methods. Qualitative results were interpreted by two technicians for Methods A and B, and by a single dIFine instrument for Method C.
K231616 - Page 9 of 17
{9}
Table 10: List of Endogenous and Exogenous Interferents
| Endogenous Interfering Substance | Maximum Concentration |
| --- | --- |
| Hemoglobin | 200 mg/mL |
| Bilirubin | 0.15 mg/mL |
| Triglycerides | 2.5 mg/mL |
| Cholesterol | 2.2 mg/mL |
| Rheumatoid Factor | 400 IU/mL |
| Intralipids | 20 mg/mL |
| Albumin | 52 mg/mL |
| Exogenous Interfering Substance | Maximum Concentration |
| --- | --- |
| Azathioprine | 0.00258 mg/mL |
| Belimumab | 8 mg/mL |
| Cyclophosphamide | 0.549 mg/mL |
| Diltiazem | 0.0009 mg/mL |
| Enalapril | 0.000819 mg/mL |
| Hydroxychloroquine | 0.24 mg/mL |
| Ibuprofen | 0.219 mg/mL |
| Intralipids | 20 mg/mL |
| Methotrexate | 1.36 mg/mL |
| Mycophenolate Mofetil | 0.048 mg/mL |
| Naproxen | 0.36 mg/mL |
| Prednisone | 0.000099 mg/mL |
| Rituximab | 2 mg/mL |
| Simvastatin | 0.000083 mg/mL |
| Voclosporin | 0.00021 mg/mL |
None of the interferents affected the expected results of any samples when read by Methods A and B. When the interferent/samples combinations were tested, Method C yielded uncertain results in several samples: 'low negative-2' sample spiked with a high concentration of cyclophosphamide, 'low positive-2' sample spiked with hydroxychloroquine, 'medium positive-2' sample spiked with a low concentration of azathioprine, and a high concentration of bilirubin, 'high positive-1' sample spiked with a high concentration of albumin, and a low concentration of triglycerides.
b. Cross-Reactivity
The analytical cross-reactivity of the assay was evaluated using 23 International Consensus on Antinuclear Antibody (ANA) Patterns (ICAP) reference samples tested at a 1:10 dilution with all three interpretation methods. The results of this study demonstrate that most of the ANA exhibited by members of the ICAP panel do not cross react with the kinetoplast of the Crithidia in the ZEUS IFA nDNA Test System except for ANA-1 (ANA Homogenous positive), ANA-7 (Anti SS-A Ro Positive), ANA-23 (Anti Rods/Rings positive) that were positive across all three interpretation methods.
K231616 - Page 10 of 17
{10}
4. Assay Reportable Range:
Not applicable
5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
a. Traceability
A recognized standard or reference material for anti-dsDNA antibodies for immunofluorescence is not available.
b. Stability
i). Open Reagent Stability
The slides must be used the same day as they are opened. All other ready-to-use reagents, except PBS, may be used until their stated expiration date.
ii). Unopened Reagent Stability
The reagents for this assay are the same as those in the predicate device. The real-time stability studies support shelf-life claim of 24 months when stored at 2–8°C.
iii). Specimen Stability
The real-time sample stability data supports sample storage at room temperature (20–25°C) for no longer than 8 hours. If testing is not performed within 8 hours, sera may be stored between 2–8°C, for no longer than 48 hours. If delay in testing is anticipated, test sera must be stored at –20°C or lower.
6. Detection Limit:
Not applicable
7. Assay Cut-Off:
Please refer to K7810178. Zeus recommends a screening dilution of 1:10 and any titers less than 1:10 are considered negative.
B Comparison Studies:
1. Method Comparison with Predicate Device:
The method comparison study was performed using clinical samples (described in Section C.1) at three independent laboratories in the U.S. Considering the interpretations recorded from all three sites for these 660 specimens, there were a total of 3,960 instances where the results of Method A versus Method B, Method A versus Method C, and Method B versus Method C were compared. Two technicians tested each sample at each site. Method A is considered as the predicate.
K231616 - Page 11 of 17
{11}
a. Method A vs Method B Agreement
Table 11: Qualitative Agreement between Method A and Method B for each Site and each Technician
| Method A vs Method B | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | 100.00% (83/83) (95.58 – 100.00) | 100.00% (577/577) (99.34 – 100.00) | 100.00% (660/660) (99.42 – 100.00) |
| | Tech 2 | 98.78% (81/82) (93.41 – 99.78) | 99.65% (576/578) (98.75 – 99.91) | 99.55% (657/660) (98.67 – 99.85) |
| Site 2 | Tech 1 | 100.00% (74/74) (95.07 – 100.00) | 99.66% (584/586) (98.76 – 99.91) | 99.70% (658/660) (98.90 – 99.92) |
| | Tech 2 | 97.44% (76/78) (91.12 – 99.29) | 100.00% (582/582) (99.34 – 100.00) | 99.70% (658/660) (98.90 – 99.92) |
| Site 3 | Tech 1 | 100.00% (80/80) (95.42 – 100.00) | 99.31% (576/580) (98.24 – 99.73) | 99.39% (656/660) (98.45 – 99.76) |
| | Tech 2 | 100.00% (80/80) (95.42 – 100.00) | 98.97% (574/580) (97.76 – 99.53) | 99.09% (654/660) (98.03 – 99.58) |
Table 12: Combined Qualitative Agreement for all Sites/all Technicians
| | Method A | | Total | |
| --- | --- | --- | --- | --- |
| | | Positive | Negative | |
| Method B | Positive | 474 | 14 | 488 |
| | Negative | 3 | 3469 | 3472 |
| | Total | 477 | 3483 | 3960 |
PPA: $99.37\%$ (474/477) $(95\% \mathrm{CI}: 98.17\%$ to $99.79\%$
NPA: $99.60\%$ (3469/3483) $(95\% \mathrm{CI}: 99.33\%$ to $99.76\%$
Overall Agreement: $99.57\%$ (3943/3960) $(95\% \mathrm{CI}: 99.31\%$ to $99.73\%)$
# b. Method A vs Method C Qualitative Comparison
Since Method C can yield an uncertain (UNC) result in addition to a positive or negative qualitative result, the agreement between methods were calculated using the UNC samples considered positive and then considered negative:
Table 13: UNC considered as Positive for each Site and each Technician
| Method A vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | 97.59% (81/83) (91.63-99.34) | 98.44% (568/577) (97.06 - 99.18) | 98.33% (649/660) (97.04 - 99.07) |
| | Tech 2 | 95.18% (79/83) (88.25-98.11) | 98.10% (567/578) (96.63 - 98.93) | 97.88% (646/660) (96.47 - 98.73) |
| Site 2 | Tech 1 | 98.65% (73/74) (92.73-99.76) | 98.46% (577/586) (97.11 - 99.19) | 98.48% (650/660) (97.23 - 99.17) |
| | Tech 2 | 97.44% (76/78) (91.13-99.29) | 98.97% (576/582) (97.77 - 99.53) | 98.79% (652/660) (97.63 - 99.38) |
K231616 - Page 12 of 17
{12}
| Method A vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 3 | Tech 1 | 97.50% (78/80) (91.34-99.31) | 98.45% (571/580) (97.08 - 99.18) | 98.33% (649/660) (97.04 - 99.07) |
| | Tech 2 | 96.25% (77/80) (89.55-98.72) | 98.28% (570/580) (96.86 - 99.06) | 98.03% (647/660) (96.66 - 98.85) |
Table 14: Combined Qualitative Agreement between Method A and Method C for all Sites/all Technicians (UNC as positives):
| | Method A | | Total | |
| --- | --- | --- | --- | --- |
| | | Positive | | Negative |
| Method C | Positive | 464 | 54 | 518 |
| | Negative | 13 | 3429 | 3442 |
| | Total | 467 | 3483 | 3960 |
PPA: 97.27% (464/467) (95% CI: 95.39% to 98.40%)
NPA: 98.45% (3429/3483) (95% CI: 97.98% to 98.81%)
Overall Agreement: 98.31% (3893/3960) (95% CI: 97.86% to 98.66%)
Table 15: UNC considered as Negative for each Site and each Technician
| Method A vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | 95.18% (79/83) (88.25 - 98.11) | 99.13% (572/577) (97.99 - 99.63) | 98.64% (651/660) (97.43 - 99.28) |
| | Tech 2 | 93.90% (77/82) (86.51 - 97.37) | 98.79% (571/578) (97.52 - 99.41) | 98.18% (648/660) (96.85 - 98.96) |
| Site 2 | Tech 1 | 91.89% (68/74) (83.42 - 96.23) | 99.83% (585/586) (99.04 - 99.97) | 98.94% (653/660) (97.83 - 99.49) |
| | Tech 2 | 88.46% (69/78) (79.50 - 93.81) | 100.00% (582/582) (99.34 - 100.00) | 98.64% (651/660) (97.43 - 99.28) |
| Site 3 | Tech 1 | 96.25% (77/80) (89.55 - 98.72) | 100.00% (580/580) (99.34 - 100.00) | 99.55% (657/660) (98.67 - 99.86) |
| | Tech 2 | 95.00% (76/80) (87.84 - 98.04) | 99.83% (579/580) (99.03 - 99.97) | 99.24% (655/660) (98.24 - 99.68) |
Table 16: Combined Qualitative Agreement between Method A and Method C for all Sites/all Technicians (UNC as Negatives):
| | Method A | | Total | |
| --- | --- | --- | --- | --- |
| | | Positive | | Negative |
| Method C | Positive | 446 | 14 | 460 |
| | Negative | 31 | 3469 | 3500 |
| | Total | 477 | 3483 | 3960 |
PPA: 93.50% (446/477) (95% CI: 90.92% to 95.38%)
NPA: 99.60% (3469/3483) (95% CI: 99.33% to 99.76%)
Overall Agreement: 98.31% (3893/3960) (95% CI: 97.86% to 98.66%)
K231616 - Page 13 of 17
{13}
c. Method B vs Method C Qualitative Comparison
Since Method C can yield an uncertain (UNC) result in addition to a positive or negative qualitative result, the agreement between methods were calculated using the UNC samples considered positive and then considered negative:
Table 17: UNC considered as Positive for each Site and each Technician
| Method B vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | 97.59% (81/83) (91.63 - 99.34) | 98.44% (568/577) (97.06 - 99.18) | 98.33% (649/660) (97.04 - 99.07) |
| | Tech 2 | 96.39% (80/83) (89.90 - 98.76) | 98.27% (567/577) (96.84 - 99.06) | 98.03% (647/660) (96.66 - 98.85) |
| Site 2 | Tech 1 | 97.37% (74/76) (90.90 - 99.28) | 98.63% (576/584) (97.32 - 99.30) | 98.48% (650/660) (97.23 - 99.17) |
| | Tech 2 | 98.68% (75/76) (92.92 - 99.77) | 98.80% (577/584) (97.55 - 99.42) | 98.79% (652/660) (97.63 - 99.38) |
| Site 3 | Tech 1 | 96.43% (81/84) (90.02 - 98.78) | 98.96% (570/576) (97.75 - 99.52) | 98.64% (651/660) (97.43 - 99.28) |
| | Tech 2 | 95.35% (82/86) (88.64 - 98.18) | 99.13% (569/574) (97.98 - 99.63) | 98.64% (651/660) (97.43 - 99.28) |
Table 18: Combined Qualitative Agreement between Method B and Method C for all Sites/all Technicians (UNC as positives):
| | Method B | | Total | |
| --- | --- | --- | --- | --- |
| | | Positive | | Negative |
| Method C | Positive | 473 | 45 | 518 |
| | Negative | 15 | 3427 | 3442 |
| | Total | 488 | 3472 | 3960 |
PPA: $96.93\%$ (473/488) $(95\% \mathrm{CI}: 94.99\%$ to $98.13\%)$
NPA: $98.70\%$ (3427/3472) $(95\% \mathrm{CI}: 98.27\%$ to $99.03\%)$
Overall Agreement: $98.48\%$ (3900/3960) $(95\% \mathrm{CI}: 98.06\%$ to $98.82\%)$
Table 19: UNC considered as Negative for each Site and each Technician
| Method B vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | 95.18% (79/83) (88.25 - 98.11) | 99.13% (572/577) (97.99 - 99.63) | 98.64% (651/660) (97.43 - 99.28) |
| | Tech 2 | 93.98% (78/83) (86.66 - 97.40) | 98.96% (571/577) (97.75 - 99.52) | 98.33% (649/660) (97.04 - 99.07) |
| Site 2 | Tech 1 | 89.47% (68/76) (80.58 - 94.57) | 99.83% (583/594) (99.04 - 99.97) | 98.64% (651/660) (97.43 - 99.28) |
| | Tech 2 | 90.79% (69/76) (82.19 - 95.47) | 100.00% (584/584) (99.35 - 100.00) | 98.94% (653/660) (97.83 - 99.49) |
K231616 - Page 14 of 17
{14}
K231616 - Page 15 of 17
| Method B vs Method C | | Positive Sample Agreement (n/N) (95% CI) | Negative Sample Agreement (n/N) (95% CI) | Total Sample Agreement (n/N) (95% CI) |
| --- | --- | --- | --- | --- |
| Site 3 | Tech 1 | 91.67% (77/84) (83.78 - 95.90) | 100.00% (576/576) (99.34 - 100.00) | 98.94% (653/660) (97.83 - 99.49) |
| | Tech 2 | 89.53% (77/86) (81.29 - 94.40) | 100.00% (574/574) (99.34 - 100.00) | 98.64% (651/660) (97.43 - 99.28) |
Table 20: Combined Qualitative Agreement between Method A and Method C for all Sites/all Technicians (UNC as Negatives):
| | Method B | | Total | |
| --- | --- | --- | --- | --- |
| | | Positive | | Negative |
| Method C | Positive | 448 | 12 | 460 |
| | Negative | 40 | 3460 | 3500 |
| | Total | 488 | 3472 | 3960 |
PPA: 91.80% (448/488) (95% CI: 89.03% to 93.92%)
NPA: 99.65% (3460/3472) (95% CI: 99.40% to 99.80%)
Overall Agreement: 98.69% (3908/3960) (95% CI: 98.28% to 98.82%)
2. Matrix Comparison:
Not applicable
C Clinical Studies:
1. Clinical Sensitivity and Clinical Specificity:
Clinically characterized specimens, described below, were tested to determine the clinical sensitivity and specificity of the assay on all three methods. The samples were acquired from commercial sources, aliquoted into three samples, then tested at three separate clinical laboratories. At all sites, the same cohort of 300 samples associated with Systemic Lupus Erythematosus (SLE) and 360 samples associated with non-SLE diseases (ANA-associated diseases and non-ANA associated diseases) were tested. The samples were tested at the screening 1:10 sample dilution then read by all three methods. At all three sites, Method A and Method B were interpreted by two different laboratory technicians.
Table 21: Clinically Characterized Samples used in the Study
| Target Disease | | n |
| --- | --- | --- |
| Systemic Lupus Erythematosus | | 300 |
| Control Diseases | | n |
| ANA-Associated Diseases | | |
| Connective Tissue Diseases | Sjögren's Syndrome | 30 |
| | Scleroderma | 20 |
| | Autoimmune Myositis | 30 |
| | Mixed Connective Tissue Disease | 20 |
| | CREST | 20 |
{15}
The clinical sensitivity was calculated at each site on SLE while specificity was calculated using the ANA associated diseases and non-ANA-associated diseases described above. The clinical sensitivity and specificity for each site and each technician is presented in the table below:
Table 22: Clinical Performance at Site 1
| Diagnostic Sensitivity and Specificity | SLE | Control Diseases |
| --- | --- | --- |
| % Sensitivity (95% CI) | % Specificity (95% CI) |
| Method A | Technician A | 26.67 (21.98 - 31.94) | 99.17 (97.58 - 99.72) |
| Technician B | 27.00 (22.29 - 32.29) | 99.72 (98.44 - 99.95) |
| Method B | Technician A | 26.67 (21.98 - 31.94) | 99.17 (97.58 - 99.72) |
| Technician B | 27.00 (22.29 - 32.29) | 99.44 (98.00 - 99.85) |
| Method C | dIFine | 27.00 (22.29 - 32.29) | 99.17 (97.58 - 99.72) |
Table 23: Clinical Performance at Site 2
| Diagnostic Sensitivity and Specificity | SLE | Control Diseases | |
| --- | --- | --- | --- |
| | | % Sensitivity (95% CI) | % Specificity (95% CI) |
| Method A | Technician A | 24.33 (19.82 - 29.49) | 99.72 (98.44 - 99.95) |
| | Technician B | 25.00 (20.44 - 30.20) | 99.17 (97.58 - 99.72) |
| Method B | Technician A | 25.00 (20.44 - 30.20) | 99.72 (98.44 - 99.95) |
| | Technician B | 24.33 (19.82 - 29.49) | 99.17 (97.58 - 99.72) |
| Method C | dIFine | 22.33 (17.99 - 27.38) | 99.17 (97.58 - 99.72) |
K231616 - Page 16 of 17
{16}
Table 24: Clinical Performance at Site 3
| Diagnostic Sensitivity and Specificity | SLE | Control Diseases | |
| --- | --- | --- | --- |
| | | % Sensitivity (95% CI) | % Specificity (95% CI) |
| Method A | Technician A | 25.33 (20.74 - 30.55) | 98.89 (97.18 - 99.57) |
| | Technician B | 25.67 (21.05 - 30.90) | 99.17 (97.58 - 99.72) |
| Method B | Technician A | 25.33 (20.74 - 30.55) | 97.78 (95.68 - 98.87) |
| | Technician B | 25.67 (21.05 - 30.90) | 97.50 (95.32 - 98.68) |
| Method C | dIFine | 24.33 (19.82 - 29.49) | 97.50 (95.32 - 98.68) |
# D Clinical Cut-Off:
Not applicable
# E Expected Values/Reference Range:
One hundred and eighty serum samples were acquired from apparently healthy donors. Samples were collected from donors within the United States. Each sample was tested at the screening dilution of 1:10 using the ZEUS IFA nDNA Test System, then scanned by ZEUS dIFine and the qualitative results (i.e., Positive, Negative or Uncertain) were determined via Methods A, B, and C. The percent positivity was determined for the 1:10 screening data. The results for the reference range are summarized below in the table:
Table 25: Reference Range Determination (N=180)
| Method | Number of Positives | % Positives | Number of Negatives | % Negatives | Number of Uncertain | % Uncertain |
| --- | --- | --- | --- | --- | --- | --- |
| A | 2 | 1.11% | 178 | 98.89% | NA | NA |
| B | 2 | 1.11% | 178 | 98.89% | NA | NA |
| C | 1 | 0.56% | 176 | 97.78% | 3 | 1.67% |
# 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 supports a substantial equivalence decision.
K231616 - Page 17 of 17
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.