Cell-Free DNA BCT is a direct-draw venous whole blood collection device intended for collection, stabilization, and transport of venous whole blood samples for use in conjunction with cell-free DNA next generation sequencing liquid biopsy assays that have been cleared or approved for use with samples collected in the Cell-Free DNA BCT device.
Device Story
Sterile, single-use, 10mL evacuated blood collection tube; contains K3EDTA anticoagulant and cell preservatives. Used by phlebotomists/clinicians for venous blood draw; tube inserted into holder/adapter; blood flows via vacuum; tube inverted 10 times for mixing. Preserves cfDNA integrity during transport to laboratory for plasma isolation and NGS liquid biopsy analysis. Enables standardized sample handling for specific NGS assays; ensures cfDNA stability for up to 8-10 days at room temperature, including shipping. Output is stabilized whole blood sample; clinical decision-making relies on downstream NGS assay results (e.g., Guardant Shield) for cancer detection/monitoring.
Clinical Evidence
Bench testing only. Evaluated cfDNA stability and variant call concordance using Guardant Shield assay. Studies included repeatability (n=59 healthy, n=29 CRC), lot-to-lot reproducibility, preservative interference, mixing variations, short-draw effects, shelf-life (up to 18 months), and shipping stability (including extreme temperature cycling). Concordance measured via Positive Percent Agreement (PPA) and Negative Percent Agreement (NPA). Results confirmed device maintains cfDNA integrity for downstream NGS analysis.
Technological Characteristics
Direct-draw venous whole blood collection tube; contains chemical stabilization agents to preserve cell-free DNA; vacuum-sealed for specific draw volume; non-sterile or sterile (as per standard blood collection tube manufacturing); standalone collection device.
Indications for Use
Indicated for collection, stabilization, and transport of venous whole blood samples for use with cleared/approved cell-free DNA next-generation sequencing liquid biopsy assays. Performance established only for Guardant360 CDx and Guardant Shield assays. Not for RNA, viral, or microbial nucleic acid stabilization; not for clinical chemistry assays; not for injection into patients.
Regulatory Classification
Identification
A blood collection device for cell-free nucleic acids is a device intended for medical purposes to collect, store, transport, and handle blood specimens and to stabilize and isolate cell-free nucleic acid components prior to further testing.
Special Controls
*Classification.* Class II (special controls). The special controls for this device are:(1) Design verification and validation documentation must include appropriate design inputs and design outputs that are essential for the proper functioning of the device for its intended use, including all of its indications for use, and must include the following:
(i) Documentation demonstrating that appropriate, as determined by FDA, measures are in place (
*e.g.,* validated device design features and specifications) to ensure that users of blood collection device for cell-free nucleic acids devices are not exposed to undue risk of bloodborne pathogen exposure and operator injury during use of the device, including blood collection, transportation, and centrifugation processes.(ii) Documentation demonstrating that appropriate, as determined by FDA, measures are in place (
*e.g.,* validated device design features and specifications) to ensure that the device reproducibly and reliably collects, transports, stabilizes, and isolates cell-free nucleic acids of sufficient yield and quality suitable for downstream applications as appropriate for its intended use. At a minimum, these measures must include:(A) Data demonstrating that blood samples collected in the device have reproducible cell-free nucleic acid yields that are suitable, as determined by FDA, for downstream testing as appropriate for the intended use, including estimates of within-lot, within-device, and lot-to-lot variability;
(B) Data demonstrating that cell-free nucleic acid yields isolated from blood specimens collected into the device do not add clinically significant bias to test results obtained using the downstream application(s) described in the intended use. For devices indicated for use with multiple downstream applications, data demonstrating acceptable performance for each type of claimed use or, alternatively, an appropriate, as determined by FDA, clinical justification for why such data are not needed;
(C) Data demonstrating that the device appropriately stabilizes cell-free nucleic acids after sample collection, during storage, and during transport over the claimed shelf life of the device;
(D) Data demonstrating that samples collected in the device have minimal levels of contamination with other types of nucleic acids present in cells or cellular components, and that these levels of contamination do not interfere with downstream testing;
(E) Data from analytical or clinical studies that demonstrate that, when used as intended, the device consistently draws a blood sample volume that is within the indicated fill range;
(F) Data from analytical or clinical studies that demonstrate that, when used as intended, cell-free nucleic acid yield, stability, and quality are not significantly impacted by interference due to other parts of the device (such as reduced or excess active ingredient) or specimen collection and processing procedures (such as hemolysis, centrifugation, or mixing of blood with anticoagulant or additives); and
(G) Data from analytical studies that demonstrate that the device is suitable for its intended use across all storage and sample handling conditions described in the device labeling, including device shelf life and shipping conditions (
*e.g.,* temperature, humidity, duration).(iii) A protocol, reviewed and determined acceptable by FDA, that specifies the verification and validation activities that will be performed for anticipated device modifications to reevaluate performance claims or performance specifications. This protocol must include a process for assessing whether a modification to technology, engineering, performance, materials, specifications, or indications for use, or any combination thereof, could significantly affect the safety or effectiveness of the device. The protocol must include assessment metrics, acceptance criteria, and analytical methods for the performance testing of changes.
Submission Summary (Full Text)
{0}
FDA
U.S. FOOD & DRUG
ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY ONLY
## I Background Information:
A 510(k) Number
K231776
B Applicant
Streck, Inc.
C Proprietary and Established Names
Cell-Free DNA BCT
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| QMA | Class II | 21 CFR 862.1676 - Blood Collection Device for Cell-Free Nucleic Acids | CH - Clinical Chemistry |
## II Submission/Device Overview:
A Purpose for Submission:
Modification to an existing device
B Measurand:
Not applicable – blood collection device
C Type of Test:
Not applicable
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
K231776 - Page 2 of 10
## III Intended Use/Indications for Use:
### A Intended Use(s):
See Indications for Use below.
### B Indication(s) for Use:
Cell-Free DNA BCT is a direct-draw venous whole blood collection device intended for collection, stabilization, and transport of venous whole blood samples for use in conjunction with cell-free DNA next generation sequencing liquid biopsy assays that have been cleared or approved for use with samples collected in the Cell-Free DNA BCT device.
### C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
Performance characteristics for this device have only been established on the Guardant360 CDx assay and Guardant Shield assay.
Do not store outside of established conditions.
Do not transfer samples drawn into tubes containing other anti-coagulants and/or preservatives into Cell-Free DNA BCT.
Do not use past expiration date printed on label.
Do not use for clinical chemistry assays or assays other than liquid biopsy next-generation sequencing.
Do not use for collection of materials to be injected into patients.
Cell-Free DNA BCT is not intended for the stabilization of RNA nor is it intended for viral or microbial nucleic acids.
### D Special Instrument Requirements:
Not applicable
## IV Device/System Characteristics:
### A Device Description:
Cell-Free DNA BCT is a sterile, single use, direct-draw blood collection tube comprised of 3 components (i.e., glass tube with rubber stopper, anticoagulant, and cell preservatives). The blood collection tube is a 10mL evacuated tube manufactured with USP Type III glass containing cerium oxide (to prevent color change associated with gamma irradiation sterilization). Each tube includes 200 µL ± 10% of liquid reagent. The reagent composition includes an anticoagulant K₃EDTA and a preservative.
{2}
K231776 - Page 3 of 10
## B Principle of Operation:
The device is intended to be placed inside a tube holder or an adaptor that contains a needle designed to pierce the tube closure and allow blood to flow into the tube. Once the vein has been penetrated (using a standard blood collection needle or a blood collection set), the tube is pushed into the holder, and the blood enters the tube. Once a tube has drawn the appropriate amount of blood (10 mL), it is disengaged from the holder and inverted 10 times to mix the reagents with the blood. The specimen is then transported to the lab for plasma isolation and extraction of cfDNA.
## V Substantial Equivalence Information:
### A Predicate Device Name(s):
Cell-Free DNA BCT
### B Predicate 510(k) Number(s):
DEN200001
### C Comparison with Predicate(s):
| Device & Predicate Device(s): | K231776 | DEN200001 |
| --- | --- | --- |
| Device Trade Name | Cell-Free DNA BCT | Same |
| General Device Characteristic Similarities | | |
| Intended Use/Indications For Use | Cell-Free DNA BCT is a direct-draw venous whole blood collection device intended for collection, stabilization, and transport of venous whole blood samples for use in conjunction with cell free DNA next-generation sequencing liquid biopsy assays that have been cleared or approved for use with samples collected in the Cell-Free DNA BCT device | Same |
| Anticoagulant | K₃EDTA | Same |
| Storage temperature for empty tube | 2-30°C | Same |
| Sample preservation after draw | 18°C to 25°C for up to 7 days total, including shipping | Same |
{3}
| Device & Predicate Device(s): | K231776 | DEN200001 |
| --- | --- | --- |
| Nominal draw volume | 10 mL | Same |
| General Device Characteristic Differences | | |
| Assay | Guardant360 CDx Assay and Guardant Shield Assay | Guardant360 CDx Assay |
VI Standards/Guidance Documents Referenced:
ISO 11137-1; 2015: Sterilization of health care products-Radiation: Part 1
ISO 11137-2: 2013 Sterilization of Health Care Products-Radiation- Part 2. Establishing the Radiation Dose
CLSI EP25-A, Evaluation of Stability of In Vitro Diagnostic Reagents
VII Performance Characteristics (if/when applicable):
A Analytical Performance:
1. Precision/Reproducibility:
Repeatability
Within-lot (between tube) variability was evaluated using native venous whole blood samples collected from 59 healthy donors into six BCTs from a single Cell Free DNA BCT manufacturing lot. For each donor, the six BCTs were processed into three cfDNA replicates. Plasma was isolated from samples within 7 days of collection. The isolated plasma was then shipped to Guardant Health (GH) on dry ice and stored at -80°C until further processing. cfDNA was extracted and sequenced on the NovaSeq 6000 Sequencing System. Of the 176 samples processed, 162 samples passed quality control metrics and were assessed for concordance. Additional data from a second study was analyzed to evaluate within lot variability in minimally manipulated colorectal cancer (CRC) samples (samples generated by combining a healthy donor sample with a clinical CRC patient plasma) from 29 donors. Three BCTs per donor from a single BCT lot were analyzed for repeatability. Specimens were shipped to GH and plasma was isolated after receipt. Guardant Shield assay (P230009) results were compared across technical replicates for each sample and assessed for concordance. Variability between replicates for each patient was evaluated based on variant call agreement for somatic variants. A concordant positive call reflects detection of an identical sequencing alteration between replicates, and a discordant call reflects the presence of an alteration in one replicate and the absence of that same alteration in another replicate. Average positive agreement (APA) and average negative agreement (ANA) were calculated as follows:
K231776 - Page 4 of 10
{4}
$$
APA = \frac{\# \text{concordant positives}}{\# \text{concordant positives} + \frac{\# \text{discordant calls}}{2}}
$$
$$
ANA = \frac{\# \text{concordant negatives}}{\# \text{concordant negatives} + \frac{\# \text{discordant calls}}{2}}
$$
The results are summarized below:
| Endpoint | Healthy | CRC |
| --- | --- | --- |
| ANA | 95.4% | Not Applicable |
| APA | 60.0% | 100% |
## Reproducibility
Lot-to-lot variability was evaluated using whole blood samples collected from healthy donors and colonoscopy-diagnosed CRC positive donors for the Guardant Shield assay. Samples were drawn into three lots of BCTs (Lot A, Lot B, and Lot C) and these lots were grouped into two conditions (Condition 1 (Lot A vs Lot B) and Condition 2 (Lot A vs Lot C)). The final donor count for samples that were included in the study is as follows: Healthy Donor Condition 1 (Lot A – Lot B) consisted of 28 donors; Healthy Donor Condition 2 (Lot A – Lot C) consisted of 22 donors; CRC Donor Condition 1 (Lot A – Lot B) consisted of 27 donors; and CRC Donor Condition 2 (Lot A – Lot C) consisted of 17 donors. An additional cohort of samples (n=25) were also collected in BCT Lot D and were compared to Lot A (Lot A vs Lot D). All Cell-Free DNA BCTs were processed into plasma within 7 days after whole blood collection, frozen, and shipped for analysis using the Guardant Shield Assay. Variability between Cell-Free DNA BCT lot was evaluated based on variant call agreement for somatic variants. APA and ANA were calculated as described previously. The results are summarized below.
| Lot Comparison | ANA | APA |
| --- | --- | --- |
| Lot A to B | 89.36% | 92.06% |
| Lot A to C | 91.30% | 87.50% |
| Lot A to D | 95.65% | 96.15% |
## 2. Linearity:
Not applicable.
K231776 - Page 5 of 10
{5}
K231776 - Page 6 of 10
## 3. Analytical Specificity/Interference:
### Preservative
To validate that the preservative formulation does not interfere with the Guardant Shield assay, venous whole blood samples were collected from healthy and CRC donors. BCTs were manufactured to reflect the normal preservative formulation ("reference"), 2x Preservative A, or 2x Preservative B. Four BCTs (2 reference BCTs and 2 test BCTs at 2x A volume) were collected from 30 healthy donors and 16 CRC donors. Four BCTs (2 reference BCTs and 2 test BCTs at 2x B volume) were collected from 30 healthy donors and 14 CRC donors. Each sample was processed to plasma into individual aliquots, frozen and shipped to Guardant Health on dry ice. One aliquot from each test group and donor was processed for cfDNA extraction and sequencing. cfDNA analysis was conducted using the Guardant Shield assay. The agreement rates were calculated for each BCT preservative condition. The percentages were calculated with the combined set of Healthy and CRC donor samples with respect to the Guardant Shield assay result for the reference condition. Performance was evaluated based on variant call concordance relative to the reference Cell-Free DNA BCT. Each variant called in the reference sample was evaluated in the experimental condition samples by positive percent agreement (PPA). The total number of concordant and discordant calls for all reference positive calls in a given experimental condition was counted across patients and used to calculate PPA. For each reference - treatment sample pair, each eligible site that is negative in the reference sample was assessed for presence of a somatic call in the treatment sample via Negative Percent Agreement (NPA). The total number of concordant and discordant calls for all reference negative calls in a given experimental condition was counted across patients and used to calculate NPA. PPA and NPA were calculated as follows:
$$
PPA \left(Experimental + \mid Reference + \right) = \frac {\# concordant positive calls}{\# reference positive calls}
$$
$$
NPA \left(Experimental - \mid Reference - \right) = \frac {\# concordant negative calls}{\# reference negative calls}
$$
The results are summarized below:
| Endpoint | 2x Preservative A | 2x Preservative B |
| --- | --- | --- |
| PPA | 92.86% | 100% |
| NPA | 96.43% | 100% |
### Incomplete Mixing
The instructions for use indicate that the tube should be inverted 10 times after collection. To evaluate the impact of variations in mixing after blood collection, specimens were collected from 30 healthy donors into six BCTs (2 per condition) for the following conditions: 10 inversions (reference condition per the instructions for use), 5 inversions, and 15 inversions. In a supplemental study, samples from 24 CRC donors were collected into four BCT as follows: 2 reference BCTs with 10 inversions and 2 BCTs with either 15 inversions or 5 inversions. Plasma was isolated and frozen into aliquots that were shipped to Guardant
{6}
Health for processing. As described above under the Preservative study, PPA and NPA were used to assess variant call concordance. The results indicated that inadequate or overmixing may result in diminished performance.
| Healthy | 5 inversions | 15 inversions |
| --- | --- | --- |
| PPA | 100.0% | 100.0% |
| NPA | 96.5% | 100.0% |
| CRC | 5 inversions | 15 inversions |
| --- | --- | --- |
| PPA | 88.89% | 100.0% |
| NPA | 66.7% | 100.0% |
## Short Draw
To evaluate potential interference caused by underfilling cfDNA BCTs, four whole blood samples were collected from healthy and CRC donors. BCTs were manufactured to reflect the normal reagent (e.g., preservative and anticoagulant) formulation ("reference"), 2x volume of reagent, or 3.3X reagent. These conditions were intended to reflect 10 mL whole blood collected, 5 mL whole blood collected, and 3 mL whole blood collected. Each sample was processed to plasma in individual aliquots, frozen and shipped to Guardant Health on dry ice. For each donor, plasma isolated from the 2 reference BCTs were pooled and processed into a single assay replicate and plasma isolated from the 2 test BCTs were pooled and processed into another single assay replicate. As described above, PPA and NPA were used to assess variant call concordance.
| Endpoint | 2x Reagent (5mL whole blood) | 3.3x Reagent (3 mL whole blood) |
| --- | --- | --- |
| PPA | 82.14% | 100% |
| NPA | 87.50% | 84.62% |
## Tube Stopper
Extractable and leachable studies were performed to identify substances within the tube stopper that may interact with patient specimens and interfere with the ability of the tube to preserve cfDNA. The results support that extractables from the tube stopper are not anticipated to interfere with device performance.
4. Assay Reportable Range:
Not applicable.
5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
## Shelf-Life
To assess the impact of age (shelf-life) of Cell-Free DNA BCT on their ability to preserve a sample of whole blood for optimal performance of the Guardant Shield assay, blood was collected from healthy and CRC subjects. Four lots of BCTs were used for each health condition (healthy or CRC). For each donor, samples were collected into two reference lots of BCTs (time 0) and two test lots of BCTs (4 months, 12 months or 18 months old at study initiation). Each sample was processed to plasma, frozen as individual aliquots, and shipped to Guardant Health on dry ice. For each donor, plasma isolated from the two reference lot
K231776 - Page 7 of 10
{7}
BCTs were pooled and processed into a single assay replicate and plasma isolated from the two test lot BCTs were pooled and processed into another single test assay replicate. PPA and NPA values were calculated comparing the reference timepoint (time 0) and the respective test timepoints, 4, 12, and 18 months. The study results support that devices stored for 18 months prior to blood collection are able to maintain cfDNA concentration and integrity for use with the Guardant Shield assay.
| Comparison | PPA | NPA |
| --- | --- | --- |
| 0 to 4 | 87.88% | 95.65% |
| 0 to 12 | 93.33% | 87.50% |
| 0 to 18 | 96.15% | 95.65% |
## Impact of Tube Storage Temperature Study
To examine the equivalency of Guardant Shield assay outputs to BCT storage temperature and the stability of whole blood stored at room temperature for up to 8 days after blood draw, 60 donors were collected into 12 BCTs across two lots. Before sample collection the BCTs (aged 15 months or greater) were stored at 2°C, 22°C, or 30°C. Once collected, both healthy and minimally manipulated CRC samples were stored for 0 or 8 days at room temperature. NPA and PPA were assessed by comparing binary Guardant Shield Test results between the reference and test condition for healthy donors and minimally manipulated CRC samples. The study results support that pre-aged devices, stored at 2-30°C, are able to maintain cfDNA concentration and integrity for use with the Guardant Shield assay for up to 8 days after blood collection.
| Healthy | Day8_22°C Reference vs Day8_2C Test | Day8_22°C Reference vs Day8_30C Test | Day0 Reference vs Day8 Test |
| --- | --- | --- | --- |
| ANA | 96.4% | 96.4% | 97.0% |
| APA | 0.0% | 50.0% | 26.2% |
| CRC | Day8_22°C Reference vs Day8_2°C Test | Day8_22°C Reference vs Day8_30°C Test | Day0 Reference vs Day8 Test |
| --- | --- | --- | --- |
| ANA | Not Applicable | Not Applicable | Not Applicable |
| APA | 100.0% | 100.0% | 100.0% |
## Sample Shipping Stability
To evaluate the stability of whole blood specimens collected in cfDNA BCTs for up to 10 days across the expected range of sample transport and storage conditions of the blood collection kits, 130 healthy and 60 positive samples were collected and evaluated. Specimens were shipped overnight to Guardant for processing to plasma. Reference BCTs pertaining to T0 testing were processed on the day of receipt. The remainder of the BCTs for each donor were stored horizontally in their respective blood collection kits until a set timepoint. Post-collection BCTs were treated for 72 hours with extreme summer or winter shipping temperatures (according to ISTA 7E, approximately -10°C to 35°C) and the remaining time at ambient temperature (18-25°C). Plasma was isolated one, six, eight, or ten days after receipt of sample. For samples subjected to extreme winter shipping conditions (potentially to -10°C) and stored for up to 10 days at room temperature, PPA was 100%, and NPA was
K231776 - Page 8 of 10
{8}
100%. For samples subjected to extreme summer shipping conditions (potentially to 35°C) and stored for up to 10 days at room temperature, PPA was 100% and NPA was 96.30%.
## Whole Blood Room Temperature
A supplemental study with 15 native CRC samples was performed to demonstrate sample stability for up to 8 days at room temperature. Each donor provided blood samples in four BCTs. Two BCTs were processed to plasma on the day of sample receipt (Day 1 after sample collection) and serve as the reference condition sample (T0). The remaining 2 BCTs were stored in the BCT for 7 days (Day 8 after sample collection) at room temperature before being processed to plasma; this served as the T1 condition sample. NPA and PPA were assessed by comparing binary Guardant Shield assay results between the reference and test condition for the CRC donors.
| CRC | Reference (Day 1) vs Condition (Day 8) |
| --- | --- |
| NPA | 50.0% |
| PPA | 100.0% |
## Additional Studies
Additional studies were conducted to assess robustness to centrifugation and stopper interference. Study protocols, acceptance criteria, and results for these studies were provided and found to be acceptable.
Stopper closure assembly stability, stopper pullout force, stopper resealing, and anticoagulant effectiveness study performance was established in DEN200001.
6. Detection Limit:
Not applicable.
7. Assay Cut-Off:
Not applicable.
## B Comparison Studies:
1. Method Comparison with Predicate Device:
Not applicable. Accuracy of results from samples collected in Cell-Free DNA BCT was established within the assay's clinical study. Refer to P230009.
2. Matrix Comparison:
Not applicable; device is for whole blood only.
K231776 - Page 9 of 10
{9}
C Clinical Studies:
1. Clinical Sensitivity:
Not applicable
2. Clinical Specificity:
Not applicable
3. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable):
Refer to P230009.
D Clinical Cut-Off:
Not applicable
E Expected Values/Reference Range:
Not applicable
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.
K231776 - Page 10 of 10
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.
