K062109 · Binax, Inc. · PSZ · Nov 9, 2006 · Microbiology
Device Facts
Record ID
K062109
Device Name
BINAXNOW INFLUENZA A & B TEST
Applicant
Binax, Inc.
Product Code
PSZ · Microbiology
Decision Date
Nov 9, 2006
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.3328
Device Class
Class 2
Attributes
Real-World Evidence, Pediatric
Real-World Evidence
Submission
Device
Sponsor
RWD Sources
RWE Use Summary
Key Tags
K062109 · Nov 9, 2006
BINAXNOW INFLUENZA A & B TEST
Binax, Inc.
Archived clinical samples from physician offices, clinics, and hospitals
Retrospective clinical samples were used to evaluate the performance (sensitivity and specificity) of the BinaxNOW Influenza A & B Test compared to culture/DFA, specifically supporting the use of nasal wash/aspirate specimens.
Retrospective Study; Retrospective evaluation of frozen clinical samples
Symptomatic patients; Sample Size: 293; Number of Sites: Multiple (physician offices, clinics, and hospitals in US and Sweden)
Culture/DFA
Sensitivity and specificity for detection of Flu A and Flu B
Indications for Use
The BinaxNOW® Influenza A & B Test is an in vitro immunochromatographic assay for the qualitative detection of influenza A and B nucleoprotein antigens in nasopharyngeal (NP) swab, nasal swab, and nasal wash/aspirate specimens. It is intended to aid in the rapid differential diagnosis of influenza A and B viral infections. Negative results do not preclude influenza virus infection and should not be used as the sole basis for treatment or other management decision.
Device Story
Lateral flow immunochromatographic assay; detects influenza A and B nucleoprotein antigens in nasopharyngeal swabs, nasal swabs, and nasal wash/aspirate specimens. Device provides qualitative visual results to aid differential diagnosis of influenza infections. Used in clinical settings (physician offices, clinics, hospitals); operated by healthcare professionals. Output is visual interpretation of test lines; assists clinicians in patient management decisions. Benefits include rapid identification of viral infection.
Clinical Evidence
Prospective study (n=1183) and retrospective study (n=293) compared device to cell culture/DFA. Prospective study: 52% pediatric, 48% adult. Flu A sensitivity ranged 75-83% and specificity 93-99% depending on sample type. Flu B sensitivity ranged 43-53% and specificity 94-100%. No performance differences observed based on age, gender, or sample type.
Technological Characteristics
Lateral flow immunochromatographic assay; utilizes monoclonal antibodies immobilized on a membrane support as three distinct lines. Form factor is a book-shaped hinged cardboard test device. No external energy source required. Manual interpretation of visual lines.
Indications for Use
Indicated for symptomatic patients of all ages, including pediatric (<18 years) and adult (≥18 years) populations, presenting with influenza-like symptoms. No contraindications specified.
Regulatory Classification
Identification
An influenza virus antigen detection test system is a device intended for the qualitative detection of influenza viral antigens directly from clinical specimens in patients with signs and symptoms of respiratory infection. The test aids in the diagnosis of influenza infection and provides epidemiological information on influenza. Due to the propensity of the virus to mutate, new strains emerge over time which may potentially affect the performance of these devices. Because influenza is highly contagious and may lead to an acute respiratory tract infection causing severe illness and even death, the accuracy of these devices has serious public health implications.
Special Controls
*Classification.* Class II (special controls). The special controls for this device are:(1) The device's sensitivity and specificity performance characteristics or positive percent agreement and negative percent agreement, for each specimen type claimed in the intended use of the device, must meet one of the following two minimum clinical performance criteria:
(i) For devices evaluated as compared to an FDA-cleared nucleic acid based-test or other currently appropriate and FDA accepted comparator method other than correctly performed viral culture method:
(A) The positive percent agreement estimate for the device when testing for influenza A and influenza B must be at the point estimate of at least 80 percent with a lower bound of the 95 percent confidence interval that is greater than or equal to 70 percent.
(B) The negative percent agreement estimate for the device when testing for influenza A and influenza B must be at the point estimate of at least 95 percent with a lower bound of the 95 percent confidence interval that is greater than or equal to 90 percent.
(ii) For devices evaluated as compared to correctly performed viral culture method as the comparator method:
(A) The sensitivity estimate for the device when testing for influenza A must be at the point estimate of at least 90 percent with a lower bound of the 95 percent confidence interval that is greater than or equal to 80 percent. The sensitivity estimate for the device when testing for influenza B must be at the point estimate of at least 80 percent with a lower bound of the 95 percent confidence interval that is greater than or equal to 70 percent.
(B) The specificity estimate for the device when testing for influenza A and influenza B must be at the point estimate of at least 95 percent with a lower bound of the 95 percent confidence interval that is greater than or equal to 90 percent.
(2) When performing testing to demonstrate the device meets the requirements in paragraph (b)(1) of this section, a currently appropriate and FDA accepted comparator method must be used to establish assay performance in clinical studies.
(3) Annual analytical reactivity testing of the device must be performed with contemporary influenza strains. This annual analytical reactivity testing must meet the following criteria:
(i) The appropriate strains to be tested will be identified by FDA in consultation with the Centers for Disease Control and Prevention (CDC) and sourced from CDC or an FDA-designated source. If the annual strains are not available from CDC, FDA will identify an alternative source for obtaining the requisite strains.
(ii) The testing must be conducted according to a standardized protocol considered and determined by FDA to be acceptable and appropriate.
(iii) By July 31 of each calendar year, the results of the last 3 years of annual analytical reactivity testing must be included as part of the device's labeling. If a device has not been on the market long enough for 3 years of annual analytical reactivity testing to have been conducted since the device received marketing authorization from FDA, then the results of every annual analytical reactivity testing since the device received marketing authorization from FDA must be included. The results must be presented as part of the device's labeling in a tabular format, which includes the detailed information for each virus tested as described in the certificate of authentication, either by:
(A) Placing the results directly in the device's § 809.10(b) of this chapter compliant labeling that physically accompanies the device in a separate section of the labeling where the analytical reactivity testing data can be found; or
(B) In the device's label or in other labeling that physically accompanies the device, prominently providing a hyperlink to the manufacturer's public Web site where the analytical reactivity testing data can be found. The manufacturer's home page, as well as the primary part of the manufacturer's Web site that discusses the device, must provide a prominently placed hyperlink to the Web page containing this information and must allow unrestricted viewing access.
(4) If one of the actions listed at section 564(b)(1)(A)-(D) of the Federal Food, Drug, and Cosmetic Act occurs with respect to an influenza viral strain, or if the Secretary of Health and Human Services (HHS) determines, under section 319(a) of the Public Health Service Act, that a disease or disorder presents a public health emergency, or that a public health emergency otherwise exists, with respect to an influenza viral strain:
(i) Within 30 days from the date that FDA notifies manufacturers that characterized viral samples are available for test evaluation, the manufacturer must have testing performed on the device with those viral samples in accordance with a standardized protocol considered and determined by FDA to be acceptable and appropriate. The procedure and location of testing may depend on the nature of the emerging virus.
(ii) Within 60 days from the date that FDA notifies manufacturers that characterized viral samples are available for test evaluation and continuing until 3 years from that date, the results of the influenza emergency analytical reactivity testing, including the detailed information for the virus tested as described in the certificate of authentication, must be included as part of the device's labeling in a tabular format, either by:
(A) Placing the results directly in the device's § 809.10(b) of this chapter compliant labeling that physically accompanies the device in a separate section of the labeling where analytical reactivity testing data can be found, but separate from the annual analytical reactivity testing results; or
(B) In a section of the device's label or in other labeling that physically accompanies the device, prominently providing a hyperlink to the manufacturer's public Web site where the analytical reactivity testing data can be found. The manufacturer's home page, as well as the primary part of the manufacturer's Web site that discusses the device, must provide a prominently placed hyperlink to the Web page containing this information and must allow unrestricted viewing access.
{0}
510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION
DECISION SUMMARY
DEVICE ONLY TEMPLATE
A. 510(k) Number: K062109
B. Purpose for Submission:
Expand the Indications for Use claim to add nasal swabs specimens, modify cautions statement, include 2 additional influenza A strains in the Analytical Reactivity claim, support use of additional transport media and update the labeling in compliance with FDA Guidance
C. Analyte:
Influenza Type A nucleoprotein antigens
D. Type of Test:
Lateral flow immunochromatographic assay
E. Applicant:
Binax, Inc.
F. Proprietary and Established Names:
BinaxNOW Influenza A & B
G. Regulatory Information:
1. Regulation section:
21 CFR Part 866.3330
2. Classification:
Antigens, CF (including CF Control), Influenza virus A, B, C
3. Product Code:
GNX
4. Panel:
83 Microbiology
H. Intended Use:
1. Intended use(s):
The BinaxNOW® Influenza A & B Test is an in vitro immunochromatographic assay for the qualitative detection of influenza A and B nucleoprotein antigens in nasopharyngeal (NP) swab, nasal swab, and nasal wash/aspirate specimens. It is intended to aid in the rapid differential diagnosis of influenza A and B viral infections. Negative results do not preclude influenza virus infection and should not be used as the sole basis for treatment or other management decision.
{1}
Page 2 of 7
Caution: Assay sensitivity for nasal wash/aspirate samples was determined primarily using archived specimens.
Users may wish to establish the sensitivity of these specimens on fresh samples.
2. Indication(s) for use: NA
3. Special condition for use statement(s): Prescription use only
4. Special instrument Requirements: NA
I. Device Description:
See: http://www.fda.gov/cdrh/reviews/K041049.pdf
J. Substantial Equivalence Information:
Predicate device name(s): BinaxNOW Influenza A & B
1. Predicate K number(s): K041049
2. Comparison with predicate:
A total of 1183 prospective specimens collected from children (less than 18 years of age) and adults (18 years or older) were evaluated in the BinaxNOW® Influenza A & B Test and compared to culture/DFA. Evaluated specimens include nasopharyngeal, nasal, and throat swabs and nasal wash/aspirates collected from patients presenting with influenza-like symptoms. Forty-three percent (43%) of the population tested was male, 57% female, 52% pediatric (< 18 years), and 48% adult (≥ 18 years). No differences in test performance were observed based on patient age or gender. A/H3 and A/H1 were the predominant influenza subtypes observed during this time.
K. Standard/Guidance Document Referenced (if applicable): NA
L. Test Principle:
See: http://www.fda.gov/cdrh/reviews/K041049.pdf
M. Performance Characteristics (if/when applicable):
1. Analytical performance:
a. Precision/Reproducibility:
{2}
Page 3 of 7
A blind study of the BinaxNOW® Influenza A & B Test was conducted at 3 separate sites using panels of blind coded specimens containing negative, low positive, and moderate positive samples. Participants tested each sample multiple times on 3 different days. There was 96.8% (242/250) agreement with expected test results, with no significant differences within run (replicates tested by one operator), between run (3 different days), between sites (3 sites), or between operators (6 operators).
b. Linearity/assay reportable range: NA
c. Traceability, Stability, Expected values (controls, calibrators, or method): NA
d. Detection limit:
The BinaxNOW® test limit of detection (LoD), defined as the concentration of influenza virus that produces positive BinaxNOW® test results approximately 95% of the time, was identified by evaluating different concentrations of inactivated Flu A/Beijing and inactivated Flu B/Harbin in the BinaxNOW® test.
Twelve (12) different operators each interpreted 2 devices run at each concentration for a total of 24 determinations per level. The following results identify a concentration of 1.03 × 10² ng/ml as the LoD for Flu A/Beijing and 6.05 × 10¹ ng/ml for Flu B/Harbin.
e. Assay cut-off:
NA
Comparison studies:
f. Method comparison with predicate device:
See J.3 above
g. Matrix comparison: NA
2. Clinical studies:
BinaxNOW® Influenza A & B Test Performance vs. Cell Culture / DFA – Prospective Study
A total of 1183 prospective specimens collected from children (less than 18 years of age) and adults (18 years or older) were evaluated in the BinaxNOW® Influenza A & B Test and compared to culture/DFA. Evaluated specimens include nasopharyngeal, nasal, and throat swabs and nasal wash/aspirates
{3}
Page 4 of 7
collected from patients presenting with influenza-like symptoms. Forty-three percent (43%) of the population tested was male, 57% female, 52% pediatric (< 18 years), and 48% adult (≥ 18 years). No differences in test performance were observed based on patient age or gender. A/H3 and A/H1 were the predominant influenza subtypes observed during this time.
BinaxNOW® A & B Test performance by sample type versus cell culture/DFA, including 95% confidence intervals, is listed below.
BinaxNOW® Influenza A & B Test Performance vs. Cell Culture/DFA for Detection of Flu A
| | Test Sensitivity | | | | Test Specificity | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | +/+ | -/+ | % Sens | 95% CI | -/- | +/- | % Spec | 95% CI |
| NP Swab | 53 | 16 | 77% | 65-86% | 278 | 3 | 99% | 97-100% |
| Nasal Swab | 85 | 17 | 83% | 74-90% | 378 | 16 | 96% | 93-98% |
| Overall | 162 | 53 | 75% | 69-81% | 947 | 21 | 98% | 97-99% |
BinaxNOW® Influenza A & B Test Performance vs. Cell Culture/DFA for Detection of Flu B
| | Test Sensitivity | | | | Test Specificity | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | +/+ | -/+ | % Sens | 95% CI | -/- | +/- | % Spec | 95% CI |
| NP Swab | 2 | 2 | 50% | 9-91% | 346 | 0 | 100% | 99-100% |
| Nasal Swab | 9 | 4 | 69% | 39-90% | 481 | 2 | 100% | 98-100% |
| Overall | 13 | 17 | 43% | 26-62% | 1150 | 3 | 100% | 99-100% |
# BinaxNOW® Influenza A & B Test Performance vs. Cell Culture / DFA – Retrospective Study
A total of 293 retrospective frozen clinical samples were evaluated in the BinaxNOW® Influenza A & B Test and compared to culture/DFA. All clinical samples were collected from symptomatic patients at multiple physician offices, clinics and hospitals located in the Southern, Northeastern and Midwestern regions of the United States and from one hospital in Sweden. Fifty-three percent (53%) of the population tested was male, 47% female, 62% pediatric (<18 years) and 38% adult (≥ 18 years). Nasal wash/aspirate specimens comprised approximately 61% of the samples tested, while NP swabs represented 39%. No differences in test performance were observed based on patient age and gender or based on sample type tested.
BinaxNOW® A & B Test performance by sample type versus cell culture/DFA, including 95% confidence intervals, is listed below.
{4}
Page 5 of 7
BinaxNOW® Influenza A & B Test Performance vs. Cell Culture/DFA for Detection of Flu A
| | Test Sensitivity | | | | Test Specificity | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | +/+ | -/+ | % Sens | 95% CI | -/- | +/- | % Spec | 95% CI |
| NP Swab | 19 | 8 | 70% | 50-86% | 77 | 9 | 90% | 81-95% |
| Wash/Aspirate | 51 | 6 | 89% | 78-96% | 117 | 6 | 95% | 89-98% |
| Overall | 70 | 14 | 83% | 73-90% | 194 | 15 | 93% | 88-96% |
BinaxNOW® Influenza A & B Test Performance vs. Cell Culture/DFA for Detection of Flu B
| | Test Sensitivity | | | | Test Specificity | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | +/+ | -/+ | % Sens | 95% CI | -/- | +/- | % Spec | 95% CI |
| NP Swab | 0 | 0 | N/A | N/A | 111 | 2 | 98% | 93-100% |
| Wash/Aspirate | 8 | 7 | 53% | 27-78% | 155 | 10 | 94% | 89-97% |
| Overall | 8 | 7 | 53% | 27-78% | 266 | 12 | 96% | 92-98% |
## Analytical Sensitivity:
The BinaxNOW® test limit of detection (LOD), defined as the concentration of influenza virus that produces positive BinaxNOW® test results approximately 95% of the time, was identified by evaluating different concentrations of inactivated Flu A/Beijing and inactivated Flu B/Harbin in the BinaxNOW® test.
Twelve (12) different operators each interpreted 2 devices run at each concentration for a total of 24 determinations per level. The following results identify a concentration of $1.03 \times 10^{2} \, \mathrm{ng/ml}$ as the LOD for Flu A/Beijing and $6.05 \times 10^{1} \, \mathrm{ng/ml}$ for Flu B/Harbin.
| Influenza A/Beijing | | |
| --- | --- | --- |
| Concentration (ng/ml) | # Detected | % Detected |
| 1.03 x 10² (LOD) | 23/24 | 96 |
| 5.60 x 10¹ (Cut-off) | * | 50 |
| Influenza B/Harbin | | |
| --- | --- | --- |
| Concentration (ng/ml) | # Detected | % Detected |
| 6.05 x 10¹ (LOD) | 23/24 | 96 |
| 2.42 x 10¹ (Cut-off) | 11/24 | 46 |
{5}
Page 6 of 7
| 3.27 x 10^{1} (High Neg) | 4/24 | 17 |
| --- | --- | --- |
| True Negative | 0/24 | 0 |
| 1.51 x 10^{1} (High Neg) | 6/24 | 25 |
| --- | --- | --- |
| True Negative | 0/24 | 0 |
*Linear regression was used to calculate a line equation, which was then used to project the cutoff concentration of Flu A/Beijing.
a. Other clinical supportive data (when a and b are not applicable):
3. Clinical cut-off: NA
4. Expected values/Reference range:
The prevalence of influenza varies from year to year, with outbreaks typically occurring during the fall and winter months. The rate of positivity found in influenza testing is dependent on many factors including the method of specimen collection, the test method used, geographic location, and the disease prevalence in specific localities. Type A viruses are typically associated with most serious influenza epidemics, while Type B are typically milder. In multi-center clinical studies conducted by Binax outside the U.S. during the 2004 respiratory season and in the US during the 2004-2005 respiratory season, the average prevalence of influenza A (as determined by viral cell culture) was 18%. The average prevalence of influenza B was 3%.
{6}
Page 7 of 7
N. Conclusion:
The submitted material 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.
