The Access CEA assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of Carcinoembryonic Antigen (CEA) levels in human serum, using the Access Immunoassay Systems. CEA measured by the Access Immunoassay Systems is used as an aid in the management of cancer patients in whom changing CEA concentrations have been observed.
Device Story
Access CEA is a two-site immunoenzymatic sandwich assay using mouse monoclonal anti-CEA antibodies. It utilizes paramagnetic particles and chemiluminescent detection to quantify CEA in human serum. The assay is designed for use on Beckman Coulter's Dxl 9000 Access Immunoassay Analyzer. It is intended for clinical laboratory use by trained personnel. The device transforms patient serum samples into quantitative CEA concentration values. Healthcare providers use these results to monitor cancer patients, specifically to track changes in CEA levels over time, which aids in clinical management decisions. The assay provides a standardized, automated method for longitudinal monitoring of disease status.
Clinical Evidence
Bench testing only. Method comparison study (n=153) demonstrated high correlation with the predicate (R=1.00, slope 0.98). Imprecision studies (CLSI EP-05-A3) showed within-laboratory CVs ranging from 2.5% to 5.2% across the measuring range (0.45–865 ng/mL). Analytical performance established: LoB 0.09 ng/mL, LoD 0.1 ng/mL, and LoQ 0.2 ng/mL. Linearity confirmed across the 0.2–1,000 ng/mL range.
Technological Characteristics
Paramagnetic particle, chemiluminescent immunoassay. Two-site immunoenzymatic sandwich format using mouse monoclonal anti-CEA antibodies. Automated operation on Dxl 9000 Access Immunoassay Analyzer. Measuring range 0.2–1,000 ng/mL. Reagents stable at 2–10°C for 28 days. Uses Lumi-Phos PRO substrate.
Indications for Use
Indicated for the quantitative determination of Carcinoembryonic Antigen (CEA) levels in human serum to aid in the management of cancer patients experiencing changing CEA concentrations.
Regulatory Classification
Identification
A tumor-associated antigen immunological test system is a device that consists of reagents used to qualitatively or quantitatively measure, by immunochemical techniques, tumor-associated antigens in serum, plasma, urine, or other body fluids. This device is intended as an aid in monitoring patients for disease progress or response to therapy or for the detection of recurrent or residual disease.
Special Controls
*Classification.* Class II (special controls). Tumor markers must comply with the following special controls: (1) A guidance document entitled “Guidance Document for the Submission of Tumor Associated Antigen Premarket Notifications (510(k)s) to FDA,” and (2) voluntary assay performance standards issued by the National Committee on Clinical Laboratory Standards.
{0}
FDA U.S. FOOD & DRUG ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY ONLY
## I Background Information:
A 510(k) Number
K223921
B Applicant
Beckman Coulter, Inc.
C Proprietary and Established Names
Access CEA
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| DHX | Class II | 21 CFR 866.6010 - Tumor-Associated Antigen Immunological Test System | IM - Immunology |
## II Submission/Device Overview:
A Purpose for Submission:
Modification of the previously cleared device – the change of substrate and updated analytical measuring interval of the assay on the DxI 9000 Immunoassay Analyzer
B Measurand:
Carcinoembryonic antigen (CEA)
C Type of Test:
Quantitative, Chemiluminescent immunoassay
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
K223921 - Page 2 of 9
## III Intended Use/Indications for Use:
### A Intended Use(s):
See Indications for Use below.
### B Indication(s) for Use:
The Access CEA assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of Carcinoembryonic Antigen (CEA) levels in human serum, using the Access Immunoassay Systems. CEA measured by the Access Immunoassay Systems is used as an aid in the management of cancer patients in whom changing CEA concentrations have been observed.
### C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
The Instruction for use of the device contains the following warnings:
The concentration of CEA in a given specimen determined with different manufacturers can vary due to differences in assay methods and reagent specificity. The results reported by the laboratory to the physician must include the identity of the CEA assay used. Values obtained with different assay methods cannot be used interchangeably. If, in the course of monitoring a patient, the assay method used for determining CEA values is changed, additional sequential testing should be carried out to confirm baseline values.
### D Special Instrument Requirements:
DxI 9000 Access Immunoassay Analyzer (refer to K221225)
## IV Device/System Characteristics:
### A Device Description:
Access CEA Reagent Kit
Each Access CEA reagent kit contains two reagent packs (50 tests/pack). Each pack contains the following:
- Solid phase: Paramagnetic particles coated with mouse anti-CEA MAb, suspended in TRIS buffered bovine serum albumin (BSA), with < 0.1% sodium azide and 0.1% ProClin 300
- Diluent: Phosphate buffer, protein (bovine, murine) with < 0.1% sodium azide and 0.1% ProClin 300
- Conjugate: Mouse anti-CEA MAb bound to alkaline phosphatase (bovine), diluted in phosphate buffer, protein (bovine), < 0.1% sodium azide and 0.1% ProClin 300
{2}
Materials needed but not supplied with reagent kit
- Access CEA Calibrators: at zero and approximately 1, 10, 100, 500 and 1,000 ng/mL
- Access CEA QC controls: at approximately 3 and 300 ng/mL
- Commercial control material
- Lumi-Phos PRO substrate
- UniCel DxI Wash Buffer II
- Access CEA Diluent (optional)
The modification of the Access CEA includes:
(a) to replace substrate (Lumi-Phos 530) of the Access CEA with a new substrate (Lumi-Phos PRO)
(b) to run on DxI 9000 Access Immunoassay Analyzer with the update the analytical measuring interval of 0.2 – 1,000 ng/mL
## B Principle of Operation:
The Access CEA assay is a two-site immunoenzymatic “sandwich” assay using two mouse monoclonal anti-CEA antibodies (MAb) which react with different epitopes of CEA. A sample is added to a reaction vessel, along with the first anti-CEA MAb-alkaline phosphatase conjugate and the second anti-CEA MAb bound to paramagnetic particles. After incubation, materials bound to the solid phase are held in a magnetic field while unbound materials are washed away. Then, the chemiluminescent substrate is added to the vessel and light generated by the reaction is measured with a luminometer. The light production is directly proportional to the concentration of analyte in the sample. Analyte concentration is automatically determined from a stored calibration.
## V Substantial Equivalence Information:
### A Predicate Device Name(s):
Access CEA
### B Predicate 510(k) Number(s):
K981985
K991707
### C Comparison with Predicate(s):
K223921 - Page 3 of 9
{3}
K223921 - Page 4 of 9
| Device & Predicate Device(s): | K223921 Candidate Device | K991707 Predicate | K981985 Predicate |
| --- | --- | --- | --- |
| Device Trade Name | Access CEA | Access CEA | |
| **General Device Characteristic Similarities** | | | |
| Intended Use/Indications For Use | The Access CEA assay is a paramagnetic particle, chemiluminescent immunoassay for the quantitative determination of Carcinoembryonic Antigen (CEA) levels in human serum, using the Access Immunoassay Systems. CEA measured by the Access Immunoassay Systems is used as an aid in the management of cancer patients *in whom changing CEA concentrations have been observed*. | Same | The Access CEA Immunoassay is a paramagnetic particle chemiluminescent immunoassay for the quantitative determination of carcinoembryonic antigen (CEA) in human serum using the ACCESS Immunoassay System. CEA measured by the ACCESS CEA Immunoassay, is used as an aid in the management of cancer patients. |
| Measurand | Carcinoembryonic antigen | Same | Same |
| Sample Type | Human serum | Same | Same |
| Assay Technology | Chemiluminescent | Same | Same |
| Measurement | Quantitative | Same | Same |
| Antibody | Monoclonal (mouse) anti-human CEA antibodies | Same | Same |
| Operating Principle | One-step sandwich | Same | Same |
| Calibrators | Multi-point calibrators containing purified human CEA | Same | Same |
| QC Controls | Bi-level controls containing human CEA | Same | Same |
| Reagent Stability | Stable at 2 to 10°C for 28 days after initial use | Same | Same |
| Method | Automated | Same | Same |
| **General Device Characteristic Differences** | | | |
| Instrument | DxI 9000 Access Immunoassay Analyzer | Access 2 Immunoassay System; UniCel DxI 800 Immunoassay System; UniCel DxI 600 Immunoassay System | |
| Substrate | Lumi-Phos PRO substrate | Access Substrate | |
| Measuring Range | 0.2 – 1,000 ng/mL | 0.1 – 1,000 ng/mL | |
{4}
VI Standards/Guidance Documents Referenced:
The following Clinical and Laboratory Standards Institute (CLSI) guidelines were used:
- CLSI EP05-A3, Evaluation of Precision of Quantitative Measurement Procedures; Approved Guideline—Third Edition
- CLSI EP06 2nd Edition, Evaluation of Linearity of Quantitative, Measurement Procedures
- CLSI EP09c 3rd Edition, Measurement Procedure Comparison and Bias Estimation Using Patient Samples
- CLSI EP17-A2, Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures; Approved Guideline– Second Edition
VII Performance Characteristics (if/when applicable):
A Analytical Performance:
All results presented below met the manufacturer’s pre-determined acceptance criteria.
1. Precision/Reproducibility:
Precision testing was performed in accordance with CLSI guideline EP05-A3.
a) Within-Laboratory Precision:
The studies were performed at a single site using one lot of the modified Access CEA reagent on one DxI 9000 Access Immunoassay Analyzer. Seven levels of human serum samples were run in a minimum of three replicates per run, two runs daily over the course of 20 days, resulting a minimum of 120 datapoints for each sample. The data were analyzed for repeatability (within-run), between-run, between-day, and within-laboratory precision. The mean (ng/mL), standard deviation (SD) (ng/mL) and percent coefficient of variation (%CV) are summarized in table below.
| Sample | Mean (ng/mL) | N | Within-Run | | Between-Run | | Between-Day | | Within-Laboratory | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV |
| 1 | 0.5 | 126 | 0.0 | 3.5 | 0.0 | 1.5 | 0.0 | 2.0 | 0.0 | 4.3 |
| 2 | 5.1 | 120 | 0.1 | 2.1 | 0.0 | 0.8 | 0.1 | 1.2 | 0.1 | 2.5 |
| 3 | 11.0 | 126 | 0.2 | 1.9 | 0.3 | 2.5 | 0.2 | 1.5 | 0.4 | 3.5 |
| 4 | 89.0 | 120 | 1.9 | 2.2 | 0.8 | 0.9 | 1.1 | 1.3 | 2.4 | 2.7 |
| 5 | 153 | 120 | 2.8 | 1.8 | 1.9 | 1.3 | 2.3 | 1.5 | 4.1 | 2.7 |
| 6 | 525 | 120 | 18.5 | 3.5 | 6.6 | 1.2 | 16.0 | 3.1 | 25.3 | 4.8 |
| 7 | 865 | 120 | 22.4 | 2.6 | 5.3 | 0.6 | 38.6 | 4.5 | 45.0 | 5.2 |
K223921 - Page 5 of 9
{5}
b) Lot-to-Lot Precision:
The study was performed at a single site using three modified Access CEA with three lots of Lumi-Phos PRO Substrate, three Access CEA Calibrator lots on two instruments. Six levels of human serum samples were run in five replicates per run, one run per day for two days, resulting a total of 180 datapoints for each sample. The data were analyzed for the repeatability (within-run), between-day, between-substrate lot, between-calibrator lot and total precision. The results are summarized in table below.
| Sample | Mean (ng/mL) | N | Within-Run | | Between-Day | | Between-Substrate Lot | | Between-Calibrator Lot | | Total | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV |
| 1 | 1.0 | 177 | 0.0 | 0.0 | 0.0 | 3.3 | 0.0 | 2.1 | 0.0 | 0.0 | 0.0 | 4.1 |
| 2 | 6.0 | 180 | 0.1 | 1.9 | 0.1 | 1.8 | 0.0 | 0.3 | 0.2 | 2.6 | 0.2 | 3.9 |
| 3 | 77.0 | 180 | 1.4 | 1.8 | 1.0 | 1.3 | 1.4 | 1.8 | 1.1 | 1.5 | 2.5 | 3.3 |
| 4 | 322 | 180 | 6.1 | 1.9 | 4.8 | 1.5 | 2.3 | 0.7 | 3.7 | 1.1 | 11.0 | 3.4 |
| 5 | 609 | 180 | 12.7 | 2.1 | 13.2 | 2.2 | 11.0 | 1.8 | 4.6 | 0.7 | 22.0 | 3.6 |
| 6 | 841 | 180 | 15.6 | 1.9 | 17.7 | 2.1 | 10.5 | 1.2 | 7.0 | 0.8 | 26.7 | 3.2 |
* 3 data point were excluded due to 'instrument error'
c) Instrument-to-Instrument Reproducibility:
The study was performed using three DxI 9000 Access Immunoassay Analyzers with a single lot of the modified Access CEA reagent. Five levels of serum samples were run in five replicates per run, one run per day over the course of five days, resulting a total of 75 datapoints for each sample. The results are summarized in table below:
| Sample | Mean (ng/mL) | N | Within-Run (Repeatability) | | Between-Day | | Between-Instrument | | Reproducibility | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV |
| 1 | 5.4 | 75 | 0.1 | 1.9 | 0.1 | 1.9 | 0.0 | 0.7 | 0.2 | 2.8 |
| 2 | 95.0 | 75 | 1.8 | 1.9 | 0.8 | 0.8 | 1.5 | 1.5 | 2.4 | 2.5 |
| 3 | 215 | 75 | 5.9 | 2.7 | 2.5 | 1.1 | 2.9 | 1.3 | 7.0 | 3.3 |
| 4 | 475 | 75 | 10.8 | 2.3 | 6.6 | 1.4 | 5.1 | 1.1 | 13.6 | 2.9 |
| 5 | 828 | 75 | 22.3 | 2.7 | 47.4 | 5.7 | 0.0 | 0.0 | 52.4 | 6.3 |
2. Linearity:
Linearity of the modified Access CEA was performed in accordance with CLSI guideline EP06 2nd Edition. Two series of samples were prepared by mixing two High samples with Low sample span the 'analytical measuring interval' (AMI) of the modified Access CEA and to provide eight dilution levels. Low sample was obtained from a single human serum sample that was specifically stripped of CEA antigen to achieve a concentration below the low end of the AMI of the assay. Two High samples were pooled human serum with CEA concentrations of $38.63\mathrm{ng / mL}$ and $1042.74\mathrm{ng / mL}$ . The low sample in the two series was run in replicates of eight, and all other dilution levels were run in replicates of four, and the 'measured value' (mean CEA ng/mL) was compared to its predicted value for deviation percentage using a weighted linear regression analysis. This deviation was then compared to the allowable deviation from linearity (ADL). The results are summarized in table below.
K223921 - Page 6 of 9
{6}
| Sample | Range (ng/mL) | Slope (95% CI) | Intercept (95% CI) | R² | % Deviation* (Deviation**) |
| --- | --- | --- | --- | --- | --- |
| 1 | 0.1 – 38.6 | 0.952 (0.936 – 0.967) | 0.003 (-0.009 – 0.015) | 0.999 | -7% – 5% (0.001) |
| 2 | 0.1 – 1042.7 | 0.966 (0.950 – 0.983) | 0.003 (-0.006 – 0.012) | 0.999 | -9% – 3% (0.00003) |
| Combined 1 and 2 | 0.1 – 1042.7 | 0.959 (0.949 – 0.969) | 0.003 (-0.003 – 0.010) | 1.000 | -8% – 4% (0.000 – 0.001) |
* % Deviation from Linearity for concentrations > 2.1 ng/mL
** Deviation (ng/mL) from Linearity for concentrations ≤ 2.1 ng/mL
The data support the linearity interval from 0.1 to 1042.7 ng/mL with the deviations from linearity within ±10%. The study results support the linearity of the claimed analytical measuring interval (AMI): 0.2 – 1,000 ng/mL.
3. Assay Reportable Range:
The reportable range is the same as the claimed measuring range: 0.2 to 1,000 ng/mL.
4. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
a) Traceability:
The traceability was established in K991707.
b) Stability:
The Access CEA Reagent kit and the Lumi-Phos PRO substrate are packaged separately. The proposed change is to replace substrate (Lumi-Phos 530) of the Access CEA with a new substrate (Lumi-Phos PRO). The shelf-life claims and on-board stability claims of Lumi-Phos PRO substrate were established in K221225.
Access CEA Reagent kit stability was established in K991707. The real-time stability study was conducted to verify the stability claims for the modified Access CEA. The results support the shelf-life of 12 months for the reagent pack stored at 2-10°C.
5. Detection Capability:
CLSI guideline EP17-A2 were followed to determine the Limit of Blank (LoB), Limit of Detection (LoD) and Limit of Quantitation (LoQ) for the modified Access CEA on DxI 9000 Access Immunoassay Analyzer.
LoB:
Four blank samples (individual native serum samples from which CEA antigen was specifically depleted) were tested in five replicates per run, one run per day over three days using two modified Access CEA reagent lots (N = 60 replicates for each lot) on two DxI 9000 Access Immunoassay Analyzers. The LoB for each reagent lot was calculated using the non-parametric approach and the maximum observed LoB is taken as the reported value for
K223921 - Page 7 of 9
{7}
the measurement procedure. LoB determined using the 95% non-parametric percentile of the replicates for each of the two reagent lots was 0.09 ng/mL and 0.03 ng/mL. The claimed LoB for the modified Access CEA on DxI 9000 Access Immunoassay Analyzer is 0.09 ng/mL.
**LoD:**
Five to seven serum samples containing low levels of CEA analyte were run in nine replicates per run, one run per day over five days using three modified Access CEA reagent lots (N = minimum 225 replicates for each lot) on three DxI 9000 Access Immunoassay Analyzers. The LoD was determined by fitting the precision profile model between within-laboratory SD and concentration. LoD calculated from LoB + "SD multiplied by the 95th percentile of the standard normal distribution" resulted in 0.1 ng/mL for each reagent lot. The claimed LoD for the modified Access CEA on DxI 9000 Access Immunoassay Analyzer is 0.1 ng/mL.
**LoQ:**
Nine to eleven serum samples containing low levels of CEA analyte were run in nine replicates of per run, one run per day over five days using three modified Access CEA reagent lots (N=minimum 405 replicates for each lot) on three DxI 9000 Access Immunoassay Analyzers. A variance components model was used to estimate the within-run and within-laboratory (total) %CV for each sample on each instrument and reagent lot combination. The LoQ was set to be the CEA concentration which met the within-laboratory imprecision of 20% CV. The claimed LoQ for the modified Access CEA on DxI 9000 Access Immunoassay Analyzer is 0.2 ng/mL.
6. **Assay Cut-Off:**
Not applicable
**B Comparison Studies:**
1. **Method Comparison with Predicate Device:**
CLSI guideline EP09c 3rd Edition were followed to compare the modified Access CEA on DxI 9000 Access Immunoassay Analyzer to the predicate device, the Access CEA on the Access 2 Immunoassay System. Patient samples falling within the analytical measuring interval of the Access CEA assay were evaluated. A total of 153 deidentified serum samples were tested with three Access CEA reagent lots and three calibrator lots, using one Lumi-Phos PRO substrate lots on three DxI 9000 Access Immunoassay Analyzers (candidate) and using one Access Substrate lots on three Access 2 instruments (predicate). Two commercial quality controls were run in duplicate each day to verify the systems were in control. The comparison between paired measurements was analyzed using Passing-Bablok method by fitting the observed modified Access CEA Assay on DxI 9000 Access Immunoassay Analyzer (dependent variable, y) into a linear regression model, with the observed values from the predicate (x, predicate). The results are summarized in the following table:
| N | Range (ng/mL) | Slope (95% CI) | Intercept (95% CI) | R |
| --- | --- | --- | --- | --- |
| 153 | 0.46 – 1071 | 0.98 (0.97 – 0.99) | 0.06 (0.00 – 0.17) | 1.00 |
K223921 - Page 8 of 9
{8}
2. Matrix Comparison:
Not applicable. The assay is for serum samples only.
C Clinical Studies:
Refer to K981985
D Clinical Cut-Off:
No cutoff for CEA monitoring has been recommended.
E Expected Values/Reference Range:
Refer to K981985
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.
K223921 - Page 9 of 9
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.