K110745 · Gold Standard Diagnostics · LYR · Mar 2, 2012 · Microbiology
Device Facts
Record ID
K110745
Device Name
HELICOBACTER PYLORI ELISA IGG TEST KIT
Applicant
Gold Standard Diagnostics
Product Code
LYR · Microbiology
Decision Date
Mar 2, 2012
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.3110
Device Class
Class 1
Attributes
Real-World Evidence
Real-World Evidence
Submission
Device
Sponsor
RWD Sources
RWE Use Summary
Key Tags
K110745 · Mar 2, 2012
HELICOBACTER PYLORI ELISA IGG TEST KIT
Gold Standard Diagnostics
Routine clinical patient serum samples
A retrospective correlation study was conducted using 625 patient samples routinely tested for H. pylori to compare the performance of the subject device against a predicate ELISA assay.
Clinical correlation study; Retrospective clinical correlation study
Adult patients suspected of having H. pylori infection or with gastrointestinal symptoms; Sample Size: 625
Micro Detect Inc. Pylori Detect IgG ELISA (K973508)
Positive, negative, and overall percent agreement
Indications for Use
The Helicobacter pylori (H. pylori) ELISA IgG test kit is intended for the qualitative detection of IgG antibodies to H. pylori in human serum in the adult population. This test is intended to aid in the diagnosis of H. pylori in patients suspected of having H. pylori infection, and in patients with gastrointestinal symptoms, and is to be used in conjunction with clinical findings.
Device Story
ELISA-based in vitro diagnostic test for qualitative detection of IgG antibodies against H. pylori in human serum. Process: purified antigen-coated microtiter wells; patient serum incubation; wash steps; HRP-conjugated anti-human IgG incubation; TMB substrate addition; colorimetric reaction; stop solution; spectrophotometric measurement at 450nm. Used in clinical laboratory settings by trained technicians. Output: optical density (OD) values compared against cutoff to determine presence of antibodies. Assists clinicians in diagnosing H. pylori infection in symptomatic patients.
Clinical Evidence
Clinical correlation study performed on 625 patient samples comparing subject device to predicate. Results: 92.7% positive agreement, 94.7% negative agreement, 20% equivocal agreement; overall agreement 92.8%. Discrepant samples resolved using a third-party assay (DiaMedix). Bench testing included intra-assay (CV 0.4-9.5%) and inter-assay (CV 7.2-10.1%) precision, reproducibility, cross-reactivity (minimal with Bacillus Cereus), and interference testing (hemoglobin, bilirubin, cholesterol, triglycerides).
Technological Characteristics
Enzyme-linked immunosorbent assay (ELISA) using polystyrene microtiter plates. Reagents: HRP-labeled conjugate, TMB substrate, acid stop solution. Detection via 450nm spectrophotometer. Qualitative interpretation based on calculated units relative to cutoff control. Standardized via CLSI guidelines (EP 7-A2, EP 17-A).
Indications for Use
Indicated for qualitative detection of IgG antibodies to H. pylori in human serum in adult patients suspected of infection or presenting with gastrointestinal symptoms; used as an aid in diagnosis in conjunction with clinical findings.
Regulatory Classification
Identification
Campylobacter fetus serological reagents are devices that consist of antisera conjugated with a fluorescent dye used to identify Campylobacter fetus from clinical specimens or cultured isolates derived from clinical specimens. The identification aids in the diagnosis of diseases caused by this bacterium and provides epidemiological information on these diseases. Campylobacter fetus is a frequent cause of abortion in sheep and cattle and is sometimes responsible for endocarditis (inflammation of certain membranes of the heart) and enteritis (inflammation of the intestines) in humans.
Predicate Devices
Micro Detect, Inc. Pylori Detect IgG ELISA (K973508)
Submission Summary (Full Text)
{0}
1
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY TEMPLATE
A. 510(k) Number:
K110745
B. Purpose for Submission:
To obtain substantial equivalence determination for a new device which detect IgG antibodies to *Helicobacter pylori* (H. pylori) in human serum.
C. Measurand:
Detection of IgG Antibodies to H. pylori
D. Type of Test:
Qualitative detection of H. pylori IgG Enzyme-Linked Immunosorbent Assay (ELISA) Test Kit
E. Applicant:
Gold Standard Diagnostics
F. Proprietary and Established Names:
*Helicobacter pylori* ELISA IgG Test Kit
G. Regulatory Information:
| Product Code | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| LYR | I | 21 CFR 866.3110 | 83 - Microbiology |
H. Intended Use:
1. Intended use:
The *Helicobacter pylori* (H. pylori) ELISA IgG test kit is intended for the qualitative detection of IgG antibodies to H. pylori in human serum in the adult population. This test is intended to aid in the diagnosis of H. pylori in patients
{1}
suspected of having H. pylori infection, and in patients with gastrointestinal symptoms, and is to be used in conjunction with clinical findings.
2. Indication for use:
The Helicobacter pylori (H. pylori) ELISA IgG test kit is intended for the qualitative detection of IgG antibodies to H. pylori in human serum in the adult population. This test is intended to aid in the diagnosis of H. pylori in patients suspected of having H. pylori infection, and in patients with gastrointestinal symptoms, and is to be used in conjunction with clinical findings.
3. Special conditions for use statement(s):
For prescription use only
4. Special instrument requirements:
Spectrophotometer - 450nm
I. Device Description:
The Gold Standard Diagnostics H. pylori ELISA IgG Test is an enzyme linked immunosorbent assay where antigen is bound to microwells in polystyrene microtiter plates. A detection antibody (Conjugate) labeled with horseradish peroxidase (HRP) to detect antibodies to H. pyori in serum is added. The Substrate contains tetramethylbenzidine (TMB) which is oxygenated by the Conjugate to produce a blue end product. The Stop Solution is an acid that stops the reaction and turns the substrate yellow if antibody is present in the patient serum. The kit also includes a Wash Buffer, Diluent, a Negative Control, Positive Control, and a Cutoff Control. The controls are provided to determine if the assay is functioning properly and to determine the antibody level. The kit contains 12 x 8well antigen coated microtiter strips in a frame. The reagents are sufficient for 96 determinations.
J. Substantial Equivalence Information:
1. Predicate Device names:
Micro Detect, Inc. Pylori Detect IgG
2. Predicate K number:
K973508
{2}
3. Comparison with predicate:
| Similarities | | |
| --- | --- | --- |
| Item | Gold Standard Diagnostics Device | Predicate |
| Intended Use | Detection of H. pylori IgG antibodies as aid in the diagnosis of H. pylori infection | Same |
| Type of test | Qualitative | Same |
| Analyte | IgG antibodies to H. pylori | Same |
| Methodology | Enzyme Linked Immuno Sorbent Assay | Same |
| Matrix | Serum | Same |
| Reader/Wavelength | Spectrophotometer /450 nm | Same |
| Differences | | |
| --- | --- | --- |
| Item | Gold Standard Diagnostic Device | Predicate |
| | | |
| | | |
| Incubation time | Incubate for 30 minutes at 37°C | Incubate for 20 minutes at Room Temperature |
| Reagents | Stop Solution – Acid mixture | Stop Solution – Sulfuric Acid |
| | Wash four times with reconstituted Wash Solution | Wash three times with reconstituted Wash Solution |
| | Add 50ul of Stop Solution | Add 100ul of Stop Solution |
K. Standard/Guidance Document Referenced (if applicable):
EP 7-A2 Interference Testing in Clinical Chemistry, Approved Guideline, 2nd ed., CLSI Vol 25, No 27, 2005
EP 17-A Protocols for determination of Limits of Detection and Limits of Quantitation; Approved Guideline, CLSI Vol 24, No.34, 2004
{3}
# L. Test Principle:
The antibody searched for in the human serum forms an immune complex with the antigen coated on the microtiter-plate. Unbound immunoglobulins are removed by washing processes. The enzyme conjugate attaches to this complex. Unbound conjugate is again removed by washing processes. After adding the substrate solution (TMB), a blue dye is produced by the bound enzyme (peroxidase). The color changes to yellow when the stopping solution is added.
# M. Performance Characteristics (if/when applicable):
# 1. Analytical performance:
# a. Precision/Reproducibility:
# Precision
The intra and inter assay precision was calculated by running six patient sera (four positives and two negatives) at three different sites. The results are summarized in the table below:
| | | Sample 1 | Sample 2 | Sample 3 | Sample 4 | Sample 5 | Sample 6 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Site 1 Intra-Assay | Ave: | 0.884 | 1.022 | 0.965 | 1.302 | 0.185 | 0.168 |
| | SD: | 0.063 | 0.043 | 0.050 | 0.006 | 0.011 | 0.016 |
| | CV: | 7.2% | 4.3% | 5.2% | 0.4% | 6.0% | 9.5% |
| Site 2 Intra-Assay | Ave: | 0.935 | 0.958 | 0.817 | 1.329 | 0.147 | 0.153 |
| | SD: | 0.028 | 0.034 | 0.042 | 0.015 | 0.005 | 0.007 |
| | CV: | 3.0% | 3.6% | 5.1% | 1.1% | 3.4% | 4.3% |
| Site 3 Intra-Assay | Ave: | 1.013 | 1.093 | 0.933 | 1.476 | 0.181 | 0.177 |
| | SD: | 0.016 | 0.103 | 0.068 | 0.106 | 0.013 | 0.011 |
| | CV: | 4.5% | 9.4% | 7.3% | 7.2% | 7.0% | 6.4% |
| Inter-Assay | Ave: | 0.974 | 1.057 | 0.925 | 1.414 | 0.177 | 0.172 |
| | SD: | 0.070 | 0.098 | 0.078 | 0.114 | 0.018 | 0.014 |
| | CV: | 7.2% | 9.3% | 8.4% | 8.0% | 10.1% | 8.4% |
# Reproducibility
The reproducibility of the assay was done by testing three samples in triplicate (a high negative, low positive and a moderate positive) for five days, twice a day, at three sites with two technicians per site. The results are summarized in the table below:
{4}
| | | 5 day average: | Sample 1 | Sample 2 | Sample 3 |
| --- | --- | --- | --- | --- | --- |
| Site 1 | Tech 1 | OD: | 0.338 | 0.708 | 1.030 |
| | | SD: | 0.047 | 0.074 | 0.095 |
| | | CV: | 13.9% | 10.4% | 9.2% |
| | Tech 2 | OD: | 0.354 | 0.692 | 1.055 |
| | | SD: | 0.043 | 0.060 | 0.107 |
| | | CV: | 12.2% | 8.7% | 10.1% |
| Site 2 | Tech 1 | OD: | 0.329 | 0.693 | 1.034 |
| | | SD: | 0.044 | 0.037 | 0.050 |
| | | CV: | 13.4% | 5.4% | 4.9% |
| | Tech 2 | OD: | 0.360 | 0.693 | 1.022 |
| | | SD: | 0.052 | 0.032 | 0.049 |
| | | CV: | 14.4% | 4.7% | 4.8% |
| Site 3 | Tech 1 | OD: | 0.300 | 0.516 | 0.888 |
| | | SD: | 0.048 | 0.054 | 0.032 |
| | | CV: | 16.1% | 10.5% | 3.6% |
| | Tech 2 | OD: | 0.374 | 0.642 | 0.928 |
| | | SD: | 0.041 | 0.094 | 0.128 |
| | | CV: | 11.0% | 14.6% | 13.7% |
The results obtained for both precision and reproducibility studies were acceptable.
b. Linearity/assay reportable range:
Not applicable
c. Traceability, Stability, Expected values (controls, calibrators, or methods):
Controls
The ready to use controls serve for a qualitative determination of antibodies. Their concentration can be expressed in units. Fluctuations resulting from the test procedure can be balanced with this calculation method to produce a high reproducibility. Use the means of the OD values for calculation of the units (see calculation of units below).
Test function control:
a) The OD of the blank should be < 0.150
The OD-values of the negative controls should be lower than the OD-values mentioned in the Quality Control Certificate. The OD-values of the positive
{5}
controls as well as of the cut-off controls should be above the OD-values mentioned in the Quality Control Certificate.
b) Units of the cut-off controls are defined as 10 Units. The calculated units of the positive controls should be within the ranges mentioned in the Quality Control Certificate. If those requirements (OD-values, units) are not fulfilled, the test has to be repeated.
## Calculation of Units:
The OD of the blank value (450/620nm) has to be subtracted from all other OD's.
$$
\text{Units}^* = (\text{OD controls or samples} / \text{OD of cutoff}) \times 10
$$
## Interpretation of the Test:
| Units | Interpretation |
| --- | --- |
| < 9.0 | negative |
| 9.0 – 11.0 | equivocal |
| > 11.0 | positive |
*Units are defined by the manufacturer.
i. If the measured values are above the defined equivocal range, they are considered to be positive.
ii. If the measured Unit is within the equivocal range, no significant high antibody concentration is present, the samples are considered to be equivocal. Retest the sample and if the results are again equivocal, no significant antibodies are present. For the definitive detection of the presence of $H.$ pylori it is necessary to determine the antibody concentration of two serum samples, one IgG and one IgM and in addition histology, culture and / or a urease test. A correct diagnosis based on the evaluation of a single serum sample is not possible.
iii. If the measured values are below the defined equivocal range, no measurable antigen specific antibodies are present in the samples. The samples are considered to be negative.
## d. Detection limit:
The limit of detection was determined by testing the kit controls along with a pooled positive sample and several dilutions of the pooled sample. The limit of detection was determined at a 30 fold dilution of the pooled sample giving an OD value of 0.379 with a corresponding unit value of 11.9.
6
{6}
7
# Analytical specificity:
## Cross Reactivity
An adsorption study was performed to evaluate cross reactivity. Sera with different levels of antibodies to *H. pylori* were adsorbed with *H. pylori*, and *Candida albicans*, *E. coli*, *Borrelia burgdorferi*, *Clostridium* spp., *Campylobacter*, *Bacillus*, *Enterobacter*, *Pseudomonas*, *Proteus* and two different strains of *H. influenza*. The identity of the bacteria used were identified by the ATCC and confirmed with the MALDI-TOF method. The bacteria were evaluated at a concentration of $10^{7}\mathrm{cfu/ml}$ or higher. The adsorbed samples were compared to the untreated samples and the mean percent inhibition was calculated. The results are summarized in the following table:
| Organism | Concentration (cfu/ml) | Mean percent inhibition |
| --- | --- | --- |
| Helicobacter pylori | | 96% |
| Candida albicans | 2.40x10^{7} | 1.8% |
| Escherichia coli | 6.90x10^{7} | 9.5% |
| Borrelia burgdorferi | 1.00x10^{8} | 5.5% |
| Clostridium spp. | 1.20x10^{7} | 6.0% |
| Campylobacter | 1.50x10^{9} | 4.7% |
| Bacillus Cereus | 4.40x10^{7} | 18.2% |
| Enterobacter | 1.80x10^{8} | 2.1% |
| Pseudomonas | 1.45x10^{8} | 5.1% |
| Haemophilus Influenza | 7.90x10^{7} | 3.8% |
| Proteus | 1.40x10^{8} | 4.6% |
The mean percent inhibition for *H. pylori* was $96\%$, and $1.8\%$ to $9.5\%$ with the other organisms. There seems to be a weak cross reactivity with *Bacillus*. Taking into account the normal test variation, the remaining cross reactivity with *Bacillus* may affect high negative samples close to the equivocal range. With the exception of the cross reactivity, no overall effects on the analytical specificity were seen on the Gold Standard Diagnostics *H. pylori* ELISA IgG assay. A limitation statement was included in the package insert.
{7}
8
# Interfering Substances
The effect of potential interfering substances on samples using the Gold Standard Diagnostics *H. pylori* ELISA IgG assay was evaluated. High levels of hemoglobin, bilirubin, cholesterol and triglycerides in serum samples were tested at the assay cutoff (9-11 units) in triplicate. The recommended concentrations from the guideline “Interference Testing In Clinical Chemistry” from the Clinical and Laboratory Standards Institute (18) were used (see table below). The tested substances did not affect the performance of the Gold Standard Diagnostics *H. pylori* ELISA IgG assay.
| Substance | Concentration | H. pylori concentration | Mean Percent Inhibition |
| --- | --- | --- | --- |
| Hemoglobin | 2 g/L | 9-11 units | 3% |
| Bilirubin | 342 μmol/L | 9-11 units | 11% |
| Cholesterol | 13 mmol/L | 9-11 units | 1% |
| Triglyceride | 37 mmol/L | 9-11 units | -19% |
Leukocytes, intestinal secretions or mucus, fat, and medications used to relieve diarrhea or other gastric symptoms were not tested, therefore, it is not known if these substances will interfere with the assay as they were not evaluated
## f. Assay cut-off:
### Determination of Cutoff
The cutoff was determined by testing normal sera from blood donors, clinical defined samples, proficiency samples, along with sera positive for *H. pylori*, borderline positive for *H. pylori*, and high negative for *H. pylori* for a total of 278 samples. The cutoff was determined by taking the negative average plus three standard deviations of the blood donors. The cutoff was further adjusted so that the clinically defined samples were positive, the proficiency samples met their criteria, the positive *H. pylori* sera were positive, and the *H. pylori* negative sera were negative.
## 2. Comparison studies:
### a. Method comparison with predicate device:
#### Performance Data
##### Method Comparison
The performance of the Gold Standard Diagnostics *H. pylori* ELISA IgG assay was determined by conducting a correlation study using 625 samples being
{8}
routinely tested for $H.$ pylori. No information as to the age, gender or ethnic makeup was provided, therefore, no conclusions can be made on these factors from these samples. The samples were tested on both the Gold Standard Diagnostics $H.$ pylori ELISA IgG assay and a commercially available ELISA assay as the predicate device. The results are summarized in the following table:
| | Predicate Device IgG ELISA | | | |
| --- | --- | --- | --- | --- |
| | | Positive | Equivocal | Negative |
| Gold Standard Diagnostics IgG ELISA | Positive | 203 | 3 | 13 |
| | Equivocal | 5 | 2 | 8 |
| | Negative | 11 | 5 | 375 |
| | Total | 219 | 10 | 396 |
% Positive Agreement = 92.7% (203/219) with 95% CI: 88.5% to 95.5%;
% Negative Agreement = 94.7% (375/396) with 95% CI: 92.0% to 96.5%
% Equivocal Agreement = 20% (2/10) with 95% CI: 5.7% to 51.0%
Overall Agreement = 92.8% (580/625)
The discrepant samples were further tested with a third commercially available assay. Of the eleven Gold Standard Diagnostics negative samples, and predicate device positive samples, nine samples were positive by the third assay. Of the 13 Gold Standard Diagnostics positive samples, and predicate device negative samples, one sample was equivocal, two negative, and ten samples were positive by the third assay.
## b. Matrix comparison:
### 3. Clinical studies:
a. Clinical Sensitivity:
No claim can be made for clinical sensitivity since comparison was not done to the gold standard e.g. biopsy (culture and/or histological diagnosis) or the urea breath test.
b. Clinical specificity:
Same as above
c. Other clinical supportive data (when a. and b. are not applicable):
### 4. Clinical cut-off:
Not applicable
{9}
5. Expected values/Reference range:
H. pylori is universally distributed and affect both genders, all ages and races. The prevalence of infection with H. pylori is higher in underdeveloped countries and in the communities with low standard of living and poor hygiene. Studies in asymptomatic Caucasians in the United States show that there is an increase in the prevalence of H. pylori infection with age (13-15). An expected values study was performed with the H. pylori IgG ELISA testing 625 patients from various populations. Results showed that a total of 203 patients were positive with the assay. The expected values result for the H. pylori IgG ELISA is 32%.
N. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.
O. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
10
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.