K181777 · Aggredyne, Inc. · JOZ · Mar 29, 2019 · Hematology
Device Facts
Record ID
K181777
Device Name
AggreGuide A-100 ADP
Applicant
Aggredyne, Inc.
Product Code
JOZ · Hematology
Decision Date
Mar 29, 2019
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 864.5700
Device Class
Class 2
Indications for Use
The AggreGuide A-100 ADP Assay is used with the AggreGuide A-100 instrument in non-CLIA waived physician’s office or clinical laboratory for the detection of platelet dysfunction in patients age 22 or older receiving P2Y12 antiplatelet drugs, prasugrel and ticagrelor, using 3.2% sodium citrated whole blood. The AggreGuide A-100 ADP Assay is a semi-quantitative assay. The level of platelet aggregation is determined by the platelet activity index (PAI) where values < 4.7 PAI suggest that platelet dysfunction is due to the presence of P2Y12 antiplatelet drugs, prasugrel and ticagrelor. The test results should be interpreted in conjunction with all other clinical and laboratory data available to the clinician.
Device Story
Disposable assay cartridge used with AggreGuide A-100 instrument; detects platelet dysfunction in patients on P2Y12 inhibitors. Cartridge contains preloaded freeze-dried ADP agonist. Whole blood sample introduced to cartridge; ADP induces platelet aggregation. Device quantifies aggregates via laser light scattering as they pass an optical window. Output is Platelet Activity Index (PAI); values < 4.7 PAI indicate platelet dysfunction due to drug presence. Used in physician offices or clinical labs by trained personnel. Clinicians interpret PAI alongside other laboratory data to assess patient response to antiplatelet therapy.
Clinical Evidence
Clinical study evaluated sensitivity against a truth data set of off-therapy vs. on-therapy patients. Sensitivity for prasugrel (post-loading) was 1.000 (95% CI: 0.918–1.000) and (post-maintenance) 0.907 (95% CI: 0.784–0.963). Sensitivity for ticagrelor (post-loading) was 0.906 (95% CI: 0.825–0.952) and (post-maintenance) 0.839 (95% CI: 0.770–0.890). Specificity was 0.907. Cut-off point determined via ROC/Youden analysis.
Indicated for patients age 22 or older receiving P2Y12 antiplatelet drugs (prasugrel and ticagrelor) to detect platelet dysfunction using 3.2% sodium citrated whole blood.
Regulatory Classification
Identification
An automated platelet aggregation system is a device used to determine changes in platelet shape and platelet aggregation following the addition of an aggregating reagent to a platelet-rich plasma.
{0}
1
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
A. 510(k) Number:
K181777
B. Purpose for Submission:
Clearance of a new device
C. Measurand:
Platelet activity index (PAI)
D. Type of Test:
Platelet aggregation
E. Applicant:
Aggregyne, Inc.
F. Proprietary and Established Names:
AggreGuide A-100 ADP Assay
G. Regulatory Information:
1. Regulation section:
21 CFR 864.5700, Automated platelet aggregation system
2. Classification:
Class II
3. Product code:
JOZ, System, Automated Platelet Aggregation
4. Panel:
{1}
Hematology (81)
## H. Intended Use:
### 1. Intended use(s):
The AggreGuide A-100 ADP Assay is used with the AggreGuide A-100 instrument in non-CLIA waived physician’s office or clinical laboratory for the detection of platelet dysfunction in patients age 22 or older receiving P2Y12 antiplatelet drugs, prasugrel and ticagrelor, using 3.2% sodium citrated whole blood. The AggreGuide A-100 ADP Assay is a semi-quantitative assay. The level of platelet aggregation is determined by the platelet activity index (PAI) where values < 4.7 PAI suggest that platelet dysfunction is due to the presence of P2Y12 antiplatelet drugs, prasugrel and ticagrelor. The test results should be interpreted in conjunction with all other clinical and laboratory data available to the clinician.
### 2. Indication(s) for use:
Same as Intended Use above
### 3. Special conditions for use statement(s):
- For prescription use only
- Evaluation of residual platelet activity for patients on ticagrelor therapy should be made within 3–6 hours of the most recent dose.
- Test results should be interpreted in conjunction with other clinical and laboratory data available to the clinician.
### 4. Special instrument requirements:
AggreGuide A-100 instrument
## I. Device Description:
The AggreGuide A-100 ADP Assay is used with the AggreGuide A-100 instrument and consists of disposable single-use cartridges with preloaded freeze dried adenosine-5-diphosphate (ADP). The AggreGuide A-100 instrument is a tabletop instrument that contains a touch screen for entering information and a detection chamber on the top surface that contains a laser diode and photodector to detect light scatter. The disposable single-use cartridges contain a sample well (reaction chamber) with preloaded freeze dried ADP reagent and a rotor to mix the whole blood sample with the ADP reagent.
## J. Substantial Equivalence Information:
### 1. Predicate device name(s):
{2}
VerifyNow PRUTest
2. Predicate 510(k) number(s):
K141427
3. Comparison with predicate:
| Similarities | | |
| --- | --- | --- |
| Item | Device AggreGuide A-100 ADP Assay K181777 | Predicate VerifyNow PRUTest K141427 |
| Intended Use | The AggreGuide A-100 ADP Assay is used with the AggreGuide A-100 instrument in non-CLIA waived physician's office or clinical laboratory for the detection of platelet dysfunction in patients age 22 or older receiving P2Y12 antiplatelet drugs, prasugrel and ticagrelor, using 3.2% sodium citrated whole blood. The AggreGuide A-100-ADP Assay is a semi-quantitative assay. The level of platelet aggregation is determined by the platelet activity index (PAI) where values < 4.7 PAI suggest that platelet dysfunction is due to the presence of P2Y12 antiplatelet drugs, prasugrel and ticagrelor. The test results should be interpreted in conjunction with all other clinical and laboratory data available to the clinician. | The VerifyNow PRUTest is a whole blood test used in the laboratory or point of care setting to measure the level of platelet P2Y12 receptor blockade. For in vitro diagnostic use. For professional use only. |
| Cartridge Type | Disposable single-use cartridge | Same |
| Specimen Type | Whole blood collected in 3.2% sodium citrate collection tubes | Same |
| Platelet Activation Agonist | Adenosine-5-diphosphate (ADP) | Same |
| Testing Condition | 37°C, approximately 5 minutes duration | Same |
| Specimen Collection-to- Test Time | Between 30 minutes and 4 hours | Same |
| Differences | | |
| --- | --- | --- |
| Item | Device AggreGuide A-100 ADP Assay K181777 | Predicate VerifyNow PRUTest K141427 |
| Test Principle | Whole blood is mixed with ADP. ADP activates platelets by binding | Fibrinogen-coated microparticles are used in the VerifyNow |
{3}
| Differences | | |
| --- | --- | --- |
| Item | Device
AggreGuide A-100 ADP Assay
K181777 | Predicate
VerifyNow PRUTest
K141427 |
| | to the ADP receptors on the platelet surface resulting in platelet aggregation. P2Y12 inhibitors inhibit ADP-induced platelet aggregation by blocking the P2Y12 ADP receptor on the platelet surface. | PRUTest device to bind activated platelet GP IIb/IIIa receptors. ADP is incorporated into the assay to activate platelets, and the reagent is formulated to specifically measure P2Y12-mediated platelet aggregation. When the activated platelets are exposed to the fibrinogen-coated microparticles, aggregation occurs in proportion to the number of activated platelet receptors. The VerifyNow PRUTest reports results in P2Y12 Reaction Units (PRU). |
| Reagent Composition | Adenosine-5-diphosphate (ADP) | Adenosine-5-diphosphate (ADP), Prostaglandin E1, human fibrinogen coated beads, peptide, fatty acid, buffer and preservative. |
| Detection Method | Optical, light scatter | Optical, light transmittance |
| Measurement Unit | PAI, platelet activity index
Dimensionless scalar | PRU, P2Y12 reaction units
Dimensionless scalar |
| Expected Values/
Reference Range | 4.7–9.6 PAI | 170–272 PRU |
| Reportable Range | 2.1–12.0 PAI
PAI < 2.1: the result is reported as “LOW” accompanied by a message “No Detectable Aggregation”
PAI > 12: the result is reported as “HIGH”, accompanied by a message “Aggregation is High, reliability of a PAI greater than 12 has not been established.” | |
| Energy Source | Optical: laser at 785 nm
Mechanical: rotating motor with magnet | Optical: near-infrared light source
Mechanical: reciprocating magnet |
K. Standard/Guidance Document Referenced (if applicable):
CLSI EP05-A3, Evaluation of Precision of Quantitative Measurement Procedures; Approved Guideline – Third Edition, 2014
CLSI EP25-A, Evaluation of Stability of In Vitro Diagnostic Reagents; Approved Guideline – Second Edition, 2009
{4}
CLSI EP28-A3c, Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory; Approved Guideline – Third Edition, 2010
CLSI EP17-A2, Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures; Approved Guideline – Second Edition, 2012
ISO 14971:2007, Medical devices – Application of risk management to medical devices – Second Edition, 2007
## L. Test Principle:
The AggreGuide A-100 instrument uses a laser diode and photodetector to measure light scatter. A whole blood sample is added to the reaction chamber of the AggreGuide A-100 ADP Assay cartridge. The reaction chamber contains a preloaded freeze-dried agonist, ADP, as well as a magnetically activated rotor that mixes the agonist with the whole blood sample. ADP activates platelets by binding to the ADP receptors on the platelet surface resulting in platelet aggregation. P2Y12 inhibitors inhibit ADP-induced platelet aggregation by blocking the P2Y12 ADP receptor on the platelet surface. The laser in the AggreGuide A-100 instrument emits light into the sample and platelet aggregates are detected by the observation of light scatter. The amount of platelet aggregation is proportional to the amount of platelet aggregates detected in the sample. Test results are reported as a Platelet Activity Index (PAI), where higher values are associated with more platelet activity/aggregation and lower values are associated with less platelet activity/aggregation.
## M. Performance Characteristics (if/when applicable):
### 1. Analytical performance:
#### a. Precision/Reproducibility:
The repeatability study was conducted using one instrument, with three donor samples across the reportable range including a high level sample (> 6.5 PAI), a sample near the cut-off (3.7–5.7 PAI) and a low level sample (< 3.7 PAI). For each donor, five to ten whole blood sample tubes were collected and three replicates were performed per tube on three different days, for a total of minimum of 60 measurements per donor. The SD results for the levels below 4.7 PAI and the %CV results for the levels above 4.7 PAI were found to be acceptable.
| Sample Level | Mean | Within-Run | | Between-Run | | Total | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | SD | %CV | SD | %CV | SD | %CV |
| High Level | 8.5 | 0.46 | 5.4 | 0.15 | 1.8 | 0.48 | 5.6 |
| Mid-Level | 4.6 | 0.47 | 10.2 | 0.00 | 0.0 | 0.47 | 10.2 |
| Low Level | 0.62 | 0.27 | 43.5 | 0.12 | 19.4 | 0.30 | 48.4 |
The reproducibility study was conducted with three normal donors, three lots of cartridges and three instruments where a single cartridge lot was used on each instrument. For each donor, seven to ten whole blood sample tubes were collected three observations
{5}
per tube, for multiple days with a minimum of 3 days to 20 days. Four runs were performed with five replicates per run on each instrument per tube. To demonstrate precision performance for each sample/donor, the $\% \mathrm{CV}$ and SD for each component of variability were calculated; including, within-run, between-run, between-lot and between-instrument. Due to biological variation of platelet function, between-donor and between-day variability was not calculated.
| Donor # | N | Mean | Within-Run | | Between-Lot | | Between-Instrument | | Between-Run | | Total Reproducibility | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | | SD | %CV | SD | %CV | SD | %CV | SD | %CV | SD | %CV |
| 022 | 560 | 8.89 | 0.63 | 7.4% | 0.19 | 2.2% | 0.03 | 0.4% | 0.66 | 7.7% | 0.94 | 11.0% |
| 023 | 140 | 8.69 | 0.62 | 7.1% | 0.0 | 0.0% | 0.22 | 2.6% | 0.66 | 7.6% | 1.08 | 12.4% |
| 033 | 220 | 9.17 | 0.64 | 7.0% | 0.38 | 4.1% | 0.75 | 8.1% | 0.68 | 7.4% | 0.75 | 8.1% |
# b. Linearity/assay reportable range:
Linearity is not applicable.
# Assay Reportable Range:
The numerical values of PAI are reported between the values of 2.1 and 12.0. When levels of aggregation are less than 2.1, the PAI is reported as "LOW" accompanied by a message "No Detectable Aggregation." When levels of aggregation are greater than 12.0, the PAI is reported as "HIGH", accompanied by a message "Aggregation is High, reliability of a PAI greater than 12 has not been established."
# c. Traceability, Stability, Expected values (controls, calibrators, or methods):
Traceability is not applicable.
# Shelf Life Stability:
In accordance with CLSI EP25-A, the shelf-life stability study was conducted for the AggreGuide A-100 ADP Assay cartridges using an isochronous study design with four different time points including baseline. Six donors representing high (normal), medium (normal) and low (dysfunctional due to P2Y12 inhibitor) were used to test shelf-life stability. Donor samples were tested on three different lots of aged cartridges (previously manufactured production lots) compared to one freshly manufactured lot that was less than one week old. Each donor was tested with each cartridge lot in five replicates per time point using a single instrument. Shelf-life stability was assessed in terms of measurand drift for each donor and lot. The data provided were found to be acceptable to support the 18-month shelf-life stability claim at room temperature.
# Transportation Stability:
The transportation (shipping) stability study was conducted with one cartridge lot,
{6}
one instrument and a single donor sample. The donor sample was collected in sodium citrate anticoagulant and tested under stressed (-20°C and 55°C) and unstressed (room temperature) conditions, for five runs and two replicates per run. To simulate a stressed condition, the assay cartridge was placed in a temperature control test compartment under different temperature cycles. The compartment temperature was lowered to -20°C and raised to 55°C. At each temperature, the cartridges were maintained for 60 minutes. The transition time from -20°C to 55°C was 15 minutes. The transition time from 55°C to -20°C was 25 minutes. The data provided for the transportation stability study was assessed with a linear regression analysis and was found to be acceptable.
Shipping temperature is monitored by an indicator attached to the box label which will change color if the temperature of the box exceeds 54°C.
## Sample Stability:
Sample stability was determined using whole blood samples collected from donors. The samples were stored at room temperature (20–25°C) and subsequently analyzed using one assay cartridge lot on one instrument at a minimum of eight different time points (ranging from 2 minutes to more than 6 hours from collection). The data supported a sample stability claim at room temperature for 30 minutes to 4 hours post-collection.
## Open-pouch Stability:
The AggreGuide A-100 ADP Assay cartridges are packaged in Mylar pouches. An open-pouch stability study was conducted to demonstrate stability of the cartridges after removal from the Mylar pouch. The study was conducted by testing three normal donor samples at four different time points (ranging from 1 minute to 60 minutes). Each time point consisted of one test and one control observation on the same instrument which was operated by the same user. For each time point, test results were compared to the respective baseline results. The data demonstrate that the AggreGuide A-100 ADP Assay cartridges should be used within 12 minutes of opening the Mylar pouch.
## Expected Values for Controls:
The labeling for the AggreGuide A-100 ADP Assay requires the quality control cartridge (QC2) test to be performed on the AggreGuide A-100 instrument every 30 days, or every 200 assays, whichever occurs first. The QC2 cartridge serves as a method of detecting gross failure of the AggreGuide A-100 instrument. The AggreGuide A-100 ADP Assay cannot be performed on the AggreGuide A-100 instrument unless a QC2 test has been successfully performed within the defined interval (30 days or 200 assays).
There are no manufactured external controls for AggreGuide A-100 ADP Assay. The
7
{7}
manufacturer recommends that laboratories establish their own donor reagent controls and may consider the following when establishing control donor groups:
- Normal PAI: In accordance with the specimen collection and handling instructions provided in the labeling, collect whole blood from healthy adult donors. The donor must not have taken any medication that is known to affect platelet function for at least seven days and should have prior platelet aggregation tests with results within the normal range established by the laboratory. The manufacturer recommends using donors whose PAI value is > 4.7.
- Abnormal PAI: In accordance with the specimen collection and handling instructions provided in the labeling, collect whole blood from a donor under chronic aspirin treatment. The manufacturer recommends using donors whose PAI value is < 4.7.
The manufacturer recommends performing quality control (normal and abnormal) for each new lot of AggreGuide A-100 ADP Assay cartridges. Quality control may be performed more frequently in accordance with laboratory procedures.
## Expected Values for Calibrators:
The AggreGuide A-100 ADP Assay is factory calibrated, as such, calibration is not performed by the user of the device.
## d. Detection limit:
In accordance with CLSI EP17-A2, the Limit of Blank (LoB) study was conducted by testing five K₂EDTA whole blood samples over 30 days, 30 runs per day, 2 replicates per run. Twenty-eight cartridge lots were run on one AggreGuide A-100 instrument. The LoB was determined to be 0.8 PAI.
## e. Analytical specificity:
Analytical specificity for the AggreGuide A-100 ADP Assay was determined by spiking normal whole blood samples collected in sodium citrate anticoagulant with various concentrations of interferents. The following endogenous and exogenous interfering substances were evaluated and showed no significant interference up to the specified concentration in whole blood samples collected in 3.2% sodium citrate anticoagulant tubes.
8
{8}
| Endogenous Substances | |
| --- | --- |
| Interferent | Concentration |
| Hemolysis | 600 mg/dL |
| Bilirubin | 10 mg/dL |
| Triglycerides | 361 mg/dL |
| Exogenous Substances | |
| Acetaminophen | 15.7 mg/dL |
| Aminocaproic acid | 0.90 mg/dL |
| Caffeine | 11.0 mg/dL |
| Captopril | 0.276 mg/dL |
| Catechin | 2.54 mg/dL |
| Chlorpromazine | 0.303 mg/dL |
| Cilostazol | 2.22 mg/dL |
| Cimetidine | 2.99 mg/dL |
| Dextran 40 | 2,425 mg/dL |
| Diltiazem | 0.0090 mg/dL |
| Dipyridamole | 1.01 mg/dL |
| Fish Oil | 32.0 mg/dL |
| Gabapentin | 2.7 mg/dL |
| Glipizide | 0.30 mg/dL |
| Glucosamine HCl | 0.203 mg/dL |
| Heparin | 340.5 U/dL |
| Ibuprofen | 21.8 mg/dL |
| Insulin | 0.00053 mg/dL |
| Liraglutide | 0.0168 mg/dL |
| L-Thyroxine | 5.00 mg/dL |
| Metformin | 1.20 mg/dL |
| Norfluoxitine | 0.0695 mg/dL |
| Norverapamil | 0.056 mg/dL |
| Omeprazole | 0.755 mg/dL |
| Oxypurinol | 1.36 mg/dL |
| Pravastatin | 0.0207 mg/dL |
| Prednisone | 0.121 mg/dL |
| Propranolol | 0.102 mg/dL |
| Salicylic Acid | 2.87 mg/dL |
| Sitagliptin | 0.12 mg/dL |
| Streptokinase | 40200 U/dL |
| Theophylline | 5.70 mg/dL |
| Valsartan | 1.17 mg/dL |
| Warfarin sodium | 7.50 mg/dL |
f. Assay cut-off:
{9}
The cut-off for the AggreGuide A-100 ADP Assay was determined to be 4.7 PAI.
# 2. Comparison studies:
a. Method comparison with predicate device:
Not applicable
b. Matrix comparison:
Not applicable
# 3. Clinical studies:
A clinical study was conducted at seven intended use sites in the U.S. using three instruments and six assay cartridge lots. For each study subject, whole blood was collected and tested at baseline and at each time point of the on-therapy state. Note: each study subject adhered to an aspirin regimen which included aspirin ingestion prior to the collection of the baseline sample and a daily $81\mathrm{mg}$ dose post-baseline.
In addition to the baseline measurement, samples were collected from subjects on ticagrelor therapy at the following time points to represent the respective phase of the on-therapy state: 3-6 hours (post-loading, "full-effect"), 24 hours to represent the "not full-effect" and 7 days (post-maintenance, "full-effect"). In addition to the baseline measurement, samples were collected from subjects on prasugrel at the following time points to represent the respective phase of the on-therapy state: 24 hours and 7 days (post-maintenance) which represent the "full-effect" states. All study samples were tested with the AggreGuide A-100 ADP Assay and the VerifyNow PRUTest.
Note: Although the clinical study included three study arms in which patients receiving ticagrelor, prasugrel, or clopidogrel were evaluated, the AggreGuide A-100 ADP Assay is intended to detect platelet dysfunction in patients receiving ticagrelor and patients receiving prasugrel. The data provided in the tables below represent the performance of ticagrelor and prasugrel as claimed in the Intended Use.
a. Clinical Sensitivity:
| Sensitivity Stratified by On-Therapy Status with 95% Confidence Intervals | | | | | |
| --- | --- | --- | --- | --- | --- |
| Drug Therapy | N | Status | On-therapy Status (time from dose) | AggreGuide A-100 ADP Assay (95% CI) | VerifyNow PRUTest (95% CI) |
| Prasugrel | 43 | Full-Effect | Post-Loading (24 hours) | 1.000 (0.918, 1.000) | 0.977 (0.877, 0.996) |
| | 43 | | Post-Maintenance (7 days) | 0.907 (0.784, 0.963) | 0.977 (0.877, 0.996) |
| Ticagrelor | 85 | Full-Effect | Post-Loading (3-6 hours) | 0.906 (0.825, 0.952) | 1.000 (0.957, 1.000) |
{10}
11
| Sensitivity Stratified by On-Therapy Status with 95% Confidence Intervals | | | | | |
| --- | --- | --- | --- | --- | --- |
| Drug Therapy | N | Status | On-therapy Status (time from dose) | AggreGuide A-100 ADP Assay (95% CI) | VerifyNow PRUTest (95% CI) |
| | 143 | | Post-Maintenance (7 days) | 0.839 (0.770, 0.890) | 0.986 (0.951, 0.998) |
| | 143 | Not Full-Effect | Post-Loading (24 hours) | 0.571 (0.491, 0.652) | 0.860 (0.794, 0.908) |
b. Clinical specificity:
| Specificity at Baseline/Prior to Administration of Study Drug | | | |
| --- | --- | --- | --- |
| Baseline Observation | N | AggreGuide A-100 ADP Assay (95% CI) | VerifyNow PRUTest (95% CI) |
| Prasugrel | 43 | 0.907 (0.774, 0.973) | 0.974 (0.874, 0.999) |
| Ticagrelor | 143 | 0.924 (0.867, 0.961) | 0.965 (0.921, 0.989) |
| Prasugrel and Ticagrelor | 186 | 0.919 (0.871, 0.951) | 0.968 (0.931, 0.985) |
c. Other clinical supportive data (when a. and b. are not applicable):
Not applicable
4. Clinical cut-off:
Not applicable
5. Expected values/Reference range:
The reference range for the AggreGuide A-100 ADP Assay was established in a study involving 129 healthy adult subjects (22–75 years of age) who were not receiving anti-platelet medication. The study cohort consisted of a diverse population of males and females with different ethnic backgrounds representative of the U.S. population. The reference range was determined by the one-sided lower bound of the 95% confidence interval. The lower limit is 4.54 PAI.
N. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Parts 809.10.
O. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
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.