The sponsor used a panel of 107 patient samples to perform a method comparison study (Least Squares linear regression) between the VITROS dHDL assay and the predicate Bayer Direct HDL Cholesterol II assay to demonstrate substantial equivalence.
The VITROS 5,1 FS Chemistry System is intended for use in the in vitro quantitative measurements of a variety of analytes of clinical interest, using both VITROS Chemistry Products Slides (colorimetric endpoint, rate, ion-selective electrode, and immunorate methods) and VITROS Chemistry Products MicroTip liquid reagents (spectrophotometric and spectrophotometric immunoassay methods.) The VITROS Chemistry Products dHDL Reagent Pack, VITROS Chemistry Products Calibrator Kit 10 is for in vitro diagnostic use only. The dHDL quantitatively measures HDL Cholesterol in plasma and serum. HDL cholesterol is used to evaluate the risk of developing coronary artery disease. The risk of CHD increases with lower HDL cholesterol concentrations.
Device Story
VITROS 5,1 FS Chemistry System is a fully automated, random-access clinical chemistry analyzer for high-volume diagnostic labs. It processes serum, plasma, urine, and CSF samples. The system utilizes two primary processing centers: Chemistry Slides (colorimetric, potentiometric, rate, immunorate) and MicroTip liquid reagents (spectrophotometric, immunoassay). For dHDL assays, the system performs multi-point rate measurements on patient samples. Operators interface via touchscreen/keyboard to manage worklists, calibration, and QC. The system performs automated sample identification, aspiration, incubation, and photometric analysis. Results are reported to the clinician to assess CHD risk based on HDL concentrations. The system features automated clot detection, level sensing, and bi-directional LIS connectivity.
Clinical Evidence
Bench testing only. Performance evaluated through method comparison studies using 107 patient samples, demonstrating a correlation of 0.9836 with the predicate. Precision studies (within-day and within-lab CV%) were conducted for multiple analytes (AST, CREA, Na+, PHYT, URIC, dHDL) over 10-22 days. Linearity and reportable ranges were established for all tested analytes.
Indicated for in vitro quantitative measurement of HDL cholesterol in human serum or plasma to evaluate coronary heart disease (CHD) risk. Used for clinical chemistry analysis in high-volume laboratory settings.
Regulatory Classification
Identification
A calibrator is a device intended for medical purposes for use in a test system to establish points of reference that are used in the determination of values in the measurement of substances in human specimens. (See also § 862.2 in this part.)
Special Controls
*Classification.* Class II (special controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.
{0}
510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION
DECISION SUMMARY
DEVICE AND INSTRUMENT TEMPLATE
A. 510(k) Number: K031924
B. Analyte: Vitros Chemistry products dHDL
C. Type of Test: The Vitros dHDL quantitatively measures HDL Cholesterol in plasma and serum.
D. Applicant: Neil Greenberg, Ph.D., Ortho-Clinical Diagnostics, Regulatory Affairs, 100 Indigo Creek, Rochester, NY 14626-5101
E. Proprietary and Established Names: Vitros 5,1 FS Chemistry Systems
Vitros Chemistry Products dHDL Reagent Pack, Vitros Chemistry Products Calibrator Kit 19, Vitros Chemistry Products FS Calibrator.
Common Names – Clinical Chemistry Analyzer, HDL Cholesterol assay
F. Regulatory Information:
1. Regulation section: Discrete photometric analyzer: 21 CFR 862.2160
Lipoprotein assay: 21 CFR 862.1475, Calibrator 21 CFR 862.1150
2. Classification: Analyzer and HDL Cholesterol Assay Class I, Calibrator Class II
3. Product Code: Clinical Chemistry Analyzer JJE, dHDL Cholesterol Assay LBR, Calibrators JIT
4. Panel: 75 for all
G. Intended Use:
1. Indication(s) for use: The VITROS 5,1 FS Chemistry System is intended for use in the in vitro quantitative measurements of a variety of analytes of clinical interest, using both VITROS Chemistry Products Slides (colorimetric endpoint, rate, ion-selective electrode, and immunorate methods) and VITROS Chemistry Products MicroTip liquid reagents (spectrophotometric and spectrophotometric immunoassay methods.)
The VITROS Chemistry Products dHDL Reagent Pack, VITROS Chemistry Products Calibrator Kit 10 is for in vitro diagnostic use only. The dHDL quantitatively measures HDL Cholesterol in plasma and serum. HDL cholesterol is used to evaluate the risk of developing coronary artery disease. The risk of CHD increases with lower HDL cholesterol concentrations.
{1}
Page 2 of 8
2. Special condition for use statement(s): N/A
3. Special instrument Requirements: The VITROS dHDL Chemistry pack is used with the VITROS 5,1 FS Chemistry System.
H. Device Description: The VITROS 5,1 FS Chemistry System is a random access, fully automated clinical chemistry analyzer intended for use in the in vitro determination of a variety of general chemistries, therapeutic drugs, drugs of abuse, proteins and other chemistries of clinical interest in biological fluids such as serum, plasma, urine and cerebral spinal fluid (sample type is chemistry dependant) The system has been designed for high volume clinical in vitro diagnostic applications. The analyzer operates in conjunction with reagents, calibrators and controls designed for use with the system. Major components include a command center/operator interface, a sampling center, a disposable tip processing center, the VITROS Chemistry Slide General Chemistry Center, and the VITROS MicroTip Special Chemistry processing center.
The VITROS Chemistry MicroTip range of products are combined with the VITROS 5,1 FS Chemistry System to perform the VITROS dHDL assay for HDL Cholesterol (reagent and calibrators). The VITROS Chemistry Products dHDL Reagent Pack, VITROS Chemistry Products Calibrator Kit 19 is for in vitro diagnostic use only. The dHDL quantitatively measures HDL Cholesterol in plasma and serum. HDL cholesterol is used to evaluate the risk of developing coronary artery disease. The risk of CHD increases with lower HDL cholesterol concentrations.
I. Substantial Equivalence Information:
1. Predicate device name(s): For the Analyzer – VITROS 950 Chemistry System K946090 & Bayer ADVIA 1650 Chemistry System K990346
For the dHDL assay Bayer Direct HDL Cholesterol II K982341
2. Predicate K number(s): For the Analyzer – VITROS 950 Chemistry System K946090 & Bayer ADVIA 1650 Chemistry System K990346
For the dHDL assay and calibrators - Bayer Direct HDL Cholesterol II K982341
3. Comparison with predicate:
| DEVICE | PREDICATE | |
| --- | --- | --- |
| A. Similarities | | |
| VITROS 5,1 FS Chemistry System Technology/Methodology – Colorimetric, Potentiometric, Rate, Immunorate Sample Containers – Cups, primary collection tubes Selective/Discrete testing, STAT capability, Sample ID input, Online QC, | Vitros 950 Technology/Methodology – Colorimetric, Potentiometric, Rate, Immunorate Sample Containers – Cups, primary | Bayer ADVIA Technology/Metodology – Colorimetric, Potentiometric, Rate, Immunorate Sample Containers – Cups, primary |
{2}
Page 3 of 8
| Test Panels, Bi-Directional Interface, Primary Container Sampling, Reagent Barcodes, Calibration display, TEST modes, Maintenance checks SW, Subsystem performance checks SW, Environmental Monitoring, Result Reporting, Patient results stored on board, S/W QC package, CLOT detection, Level sense (sample), Operator Interface | collection tubes Selective/Discrete testing, STAT capability, Sample ID input, Online QC, Test Panels, Bi-Directional Interface, Primary Container Sampling, Reagent Barcodes, Calibration display, TEST modes, Maintenance checks SW, Subsystem performance checks SW, Environmental Monitoring, Result Reporting, Patient results stored on board, S/W QC package, CLOT detection, Level sense (sample), Operator Interface | collection tubes Selective/Discrete testing, STAT capability, Sample ID input, Online QC, Test Panels, Bi-Directional Interface, Primary Container Sampling, Reagent Barcodes, Calibration display, TEST modes, Maintenance checks SW, Subsystem performance checks SW, Environmental Monitoring, Result Reporting, Patient results stored on board, S/W QC package, CLOT detection, Level sense (sample), Operator Interface |
| --- | --- | --- |
| VITROS dHDL Assay | | |
| Basic Principle, Reagents, sample type, incubation temp. | | |
| **B. Differences** | | |
| Throughput 845 tests per hour | 752 t.p.h. | 1650 t.p.h. |
| Throughput 370 samples per hour | 300 s.p.h. | 1200 s.p.h. |
| Technology Chemistry slides – Colorimetric, Poteniometric, Rate, and Immunorate. Micro tip reagents | N/A | N/A |
| Load Master Curve Info – CD-ROM | 3.5” diskette | User config |
| Load Assay Protocol info – CD-ROM | User config | User Config |
| Display Cal curve – No | No | Yes |
| Electrolyte Reference fluid – 800 10ul drops | 800 10ul drops | None |
| On-Analyzer sample capacity 150 routine, 10 STAT | 40 | 200 |
{3}
Page 4 of 8
| Sample Status screen – No | Yes | Yes |
| --- | --- | --- |
| Sample usage per assay (assay dependant) – 2 – 12ul | 5 – 11ul | 2 – 30ul |
| Waste Capacity - Slides 3,780, 1,145
VersaTips, 140 Vitros FS MicroTips, 3
MicroTip trays, 100 Vitros FS cuvettes | None stated | Requires drain with 6.8 gph flow rate. |
| Minimum Sample volume (worst case) – Microsample cup 30ul + 11ul per assay for up to 6 assays, 100ul + 11 per assay for 7 or more assays | Same | Sample cup 30ul + 50ul per assay |
| VITROS dHDL assay | | |
| Test type – Multi point rate | | 2 – point rate |
| Instrumentation – Vitros 5,1 FS Chemistry System | | Bayer ADVIA 1650 |
| Sample volume – 3ul | | 5ul |
| Incubation time – Inc 1 = 5minutes, Inc 2 = 3.2 minutes | | 10 minutes |
J. Standard/Guidance Document Referenced (if applicable) The standards used in this submission were, NCCLS Guidelines EP5A, NCCLS Guidelines EP9-A, NCCLS protocol EP6, NCCLS protocol EP6, and NCCLS protocol EP7-P.
K. Test Principle: For the Vitro FS 5,1 Chemistry System – Chemistry Slides – Colorimetric, Poteniometric, Rate & Immunorate. MicroTip Reagents – Endpoint Colorimetric and Rate, Enzyme-coupled Immunoassay, Homogenous Turbidometric Immunoassay, Enhanced Latex ImmunoAssay
L. Performance Characteristics (if/when applicable):
1. Analytical performance:
a. Precision/Reproducibility: The precision of Analyte Measurements on the VITROS 5,1 FS was determined using VITROS Chemistry Product Slides at various levels of activity and concentrations for a set of marker assays. The evaluation was performed on the VITROS 5,1 FS Chemistry system over a period of 10 days using one lot of slides and one instrument for each chemistry evaluated. Data were analyzed by Analysis of Variance (ANOVA). Below is a summary of the products included in the precision studies.
VITROS 5,1 FS versus VITROS 950
Analyte Slope Intercept R
{4}
| AST | 1.00 | -5.92 | 0.996 |
| --- | --- | --- | --- |
| CREA | 0.04 | 0.04 | 1.00 |
| Phenytoin | 1.02 | -0.09 | 0.997 |
| Na | 1.00 | -0.82 | 0.999 |
| URIC | 1.00 | 0.02 | 1.000 |
The comparison of the dHDL was made against a commercially available assay, Bayer Direct HDL Cholesterol II, with human patient samples. The method followed was NCCLS Guidelines EP9-A ("Method Comparison and Bias Estimation Using Patient Samples") a total of 107 serum samples were assayed using the VITROS dHDL and the Bayer ADVIA 1650 system. Samples were run in triplicate on the VITROS and a single assay was performed on each sample for the Bayer ADVIA.
Vitros dHDL Assay versus Bayer ADVIA 1650
Slope
0.974
Intercept
-0.0682
R
0.984
b. Linearity/assay reportable range: The linearity over the reportable (dynamic) range of the selected marker VITROS Chemistry slides was evaluated according to NCCLS protocol EP6. Two serum test pools were prepared with analyte concentrations near the extreme of reportable range. The reportable range of each assay is as follows - AST 3 - 750UL, CREATININE - 0.05 - 14.0mg/dl, Phenytoin 3.00 - 40.00umol/L, Sodium 75.0 - 250.0mmol/L, Uric Acid 0.50-17.00mg/dl Additionally a linearity was performed on the Vitros dHDL assay. Linearity was determined by diluting a concentrated high HDLC linearity fluid in saline, and samples were analyzed on the VITROS 5,1 FS system. The systems responses were plotted against the percent high pool and against calculated values to assess the degree to which the plotted curved conforms to a straight line. The equation of the line established was $y = 2.2744x - 1.7229$ with an R squared of 0.9985 this equation was used to plot regression concentrations. The data presented defines a linear range of the dHDL assay when run on the VITROS dHDL and established a reportable range of 2.66 to 135.64. This data supports a linearity claim of $3.0 - 100\mathrm{mg / dl}$ .
c. Traceability (controls, calibrators, or method): Aliquots of sample are analyzed with an internal reference method for the particular analyte.
d. Detection limit (functional sensitivity: Detection limits were determined by measurement of the linearity range of selected marker slides. The linearity was evaluated according to NCCLS protocol EP6. Two serum pools were prepared near the extremes of reportable range, the pools were them mixed to give additional pools at intermediate
{5}
Page 6 of 8
concentrations. Visual analysis determined the assay is linear across the range tested establishing a range and cutoff as listed for each marker assay. AST 3 – 750UL, CREATININE – 0.05 – 14.0mg/dl, Phenytoin 3.00 – 40.00umol/L, Sodium 75.0 – 250.0mmol/L, Uric Acid 0.50-17.00mg/dl Additionally a linearity was performed on the Vitros dHDL assay. Linearity was determined by diluting a concentrated high HDLC linearity fluid in saline, and samples were analyzed on the VITROS 5,1 FS system. The systems responses were plotted against the percent high pool and against calculated values to assess the degree to which the plotted curved conforms to a straight line. The equation of the line established was $y = 2.2744x - 1.7229$ with an R squared of 0.9985 this equation was used to plot regression concentrations. The data presented defines a linear range of the dHDL assay when run on the VITROS dHDL and established a reportable range of 2.66 to 135.64. This data supports a linearity claim of $3.0 - 100\mathrm{mg / dl}$.
e. Analytical specificity: Interferent studies were performed to determine the effect of levels of endogenous interferences on the performance of the VITROS Chemistry Products dHDL Reagent Pack using methods based on the NCCLS protocol EP7-P
f. Assay cut-off: Visual analysis determined the assay is linear across the range tested establishing a range and cutoff as listed for each marker assay. AST 3 – 750UL, CREATININE – 0.05 – 14.0mg/dl, Phenytoin 3.00 – 40.00umol/L, Sodium 75.0 – 250.0mmol/L, Uric Acid 0.50-17.00mg/dl and 3mg/dl and 100mg/dl for dHDL.
2. Comparison studies:
a. Method comparison with predicate device: The method comparison study used for the dHDL Cholesterol method followed NCCLS Guidelines EP9-A (Method Comparison and Bias Estimation Using Patient Samples).
b. Matrix comparison: A total of 107 human serum samples were assayed using the VITROS Chemistry products dHDL and the Bayer Direct HDL Cholesterol II assay. Single replicates of each sample were assayed on the Bayer ADVIA 1650 Chemistry System. The relationship between the two methods was determined by the least squares linear regression with patient samples form the VITROS dHDL as the "y" variable and those form the comparison device as the "x" variable. The relationship between the two assays determined by least square regression was VITROS dHDL Assay = 0.9472x X-0.0682 mg/dl with a correlation coefficient of 0.9836 where X equals the Bayer Direct HDL Cholesterol II assay.
3. Clinical studies:
{6}
Page 7 of 8
a. Clinical sensitivity: N/A
b. Clinical specificity: N/A
4. Clinical cut-off: N/A
5. Expected values/Reference range: The expected values and the reference range of the VITROS dHDL cholesterol follow the guidelines of the CDC/NIH National Cholesterol Education Program (NCEP) Recommendations for Measurements of High Density Lipoprotein Cholesterol.
M. Instrument Name: VITROS 5,1 FS Chemistry System
N. System Descriptions:
1. Modes of Operation: random access, batch and STAT modes
2. Software: The VITROS 5,1 FS Chemistry System utilizes the QNX Neutrino v 6.1x or later operating system. It relies upon a touch screen or keyboard for user interface. Communications with external systems are run via Ethernet connection. Sample programming may be made from a Laboratory Information System to the analyzer via this connection.
FDA has reviewed applicant's Hazard Analysis and software development processes for this line of product types: Yes ☐ X ☑ or No ☐
3. Sample Identification: Samples are identified via barcode or rack positioning.
4. Specimen Sampling and Handling: Samples consist of Plasma, serum, urine or cerebral spinal fluid depending upon analyte.
5. Assay Types: For the Vitro FS 5,1 Chemistry System – Chemistry Slides – Colorimetric, Poteniometric, Rate & Immunorate. MicroTip Reagents – Endpoint Colorimetric and Rate, Enzyme-coupled Immunoassay, Homogenous Turbidometric Immunoassay, Enhanced Latex ImmunoAssay
6. Reaction Types: For the Vitro FS 5,1 Chemistry System – Chemistry Slides – Colorimetric, Poteniometric, Rate & Immunorate. MicroTip Reagents – Endpoint Colorimetric and Rate, Enzyme-coupled Immunoassay, Homogenous Turbidometric Immunoassay, Enhanced Latex ImmunoAssay
7. Calibration: Calibrators are prepared from human serum and urine with specially prepared constituents added. Calibrators are placed upon the instrument, the VITROS 5,1 FS Chemistry system generates a worklist defining
{7}
Page 8 of 8
which calibrator to use and where to place the calibrator. After the test is run, the system performs on-line calibration math via its internal processor and if successful (i.e. all pre-defined quality parameters stored in memory are met), an updated calibration curve is saved for each analyte calibrated.
8. **Quality Control**: Quality Control is managed via specially prepared human based sera and urine with specifically assayed values to establish an expected range of acceptable values.
O. Other Supportive Instrument Performance Characteristics Data Not Covered In The "L. Performance Characteristics" Section Of The SE Determination Decision Summary.
P. **Conclusion**: I recommend that the Vitros Chemistry Products 5,1; Vitros Chemistry products dHDL; Vitros Chemistry products Calibrator kit 19 & Vitros Chemistry products FS Calibrator are substantially equivalent to their respective predicate devices.
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.