Human head and body hair samples; Sample Size: 610
Predicate device (Radioimmunoassay)
Qualitative detection of cannabinoids; discordance rate
Indications for Use
The Psychemedics Microplate EIA for Cannabinoids in Hair is an enzyme immunoassay (EIA) for the preliminary qualitative detection of cannabinoids in human head and body hair samples using a 11-nor-9-Carboxy- Δ -THC calibrator at 10 pg/10 mg hair cutoff for the purpose of identifying marijuana use. This is an in vitro diagnostic device intended exclusively for Psychemedics use only and is not intended for sale to anyone. The Psychemedics Chemiluminscent Microplate EIA for Cannabinoids assay provides only a preliminary analytical test result. A more specific alternative chemical method must be used in order to obtain a confirmed analytical result. Gas Chromatography/Mass Spectrometry/Mass Spectrometry (GC/MS/MS) is the preferred confirmatory method. Clinical consideration and professional judgment should be exercised with any drug of abuse test result, particularly when the preliminary result is positive.
Device Story
Device is an enzyme immunoassay (EIA) for preliminary detection of cannabinoids in human hair. Process: hair samples undergo pre-analytical treatment to recover cannabinoids; extract incubated with primary antibody; mixture transferred to antigen-coated microplate wells; secondary antibody conjugated to HRP added; chemiluminescent substrate added. Output: relative light units (RLU) measured via microplate reader (450/650 nm). RLU inversely proportional to cannabinoid concentration. Used in Psychemedics laboratory facilities. Positive screening results require confirmation via GC-MS/MS (triple stage quadrupole, negative chemical ionization mode). Output assists clinicians in identifying marijuana use; requires professional judgment for interpretation.
Clinical Evidence
No clinical studies performed. Analytical performance established via bench testing. Precision evaluated with 15 replicates (intra-assay) and 75 replicates (inter-assay) at concentrations ranging from -100% to +100% of cutoff. Cross-reactivity tested against structurally related and unrelated compounds. Cosmetic treatment study (bleaching, dyeing, etc.) on 60 positive and 100 negative samples showed no false negatives or false positives. Environmental contamination study confirmed that screening method can be positive due to external smoke, necessitating GC-MS/MS confirmation.
Technological Characteristics
Enzyme immunoassay (EIA) using chemiluminescent detection. Components: white microplate coated with 11-nor-9-carboxy-delta-8-tetrahydrocannabinol-BSA conjugate, polyclonal rabbit anti-cannabinoid antibody, goat anti-rabbit HRP-conjugated secondary antibody, and chemiluminescent substrate. Requires microplate washer and reader. Acidic extraction method. Cutoff: 10 pg/10 mg hair.
Indications for Use
Indicated for preliminary qualitative detection of cannabinoids in human head and body hair samples to identify marijuana use. Uses 11-nor-9-Carboxy-Δ9-THC calibrator at 10 pg/10 mg hair cutoff. Intended for professional use only; not for over-the-counter sale.
Regulatory Classification
Identification
A cannabinoid test system is a device intended to measure any of the cannabinoids, hallucinogenic compounds endogenous to marihuana, in serum, plasma, saliva, and urine. Cannabinoid compounds include delta-9-tetrahydrocannabinol, cannabidiol, cannabinol, and cannabichromene. Measurements obtained by this device are used in the diagnosis and treatment of cannabinoid use or abuse and in monitoring levels of cannabinoids during clinical investigational use.
Special Controls
*Classification.* Class II (special controls). A cannabinoid test system is not exempt if it is intended for any use other than employment or insurance testing or is intended for Federal drug testing programs. The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9, provided the test system is intended for employment and insurance testing and includes a statement in the labeling that the device is intended solely for use in employment and insurance testing, and does not include devices intended for Federal drug testing programs (*e.g.,* programs run by the Substance Abuse and Mental Health Services Administration (SAMHSA), the Department of Transportation (DOT), and the U.S. military).
Predicate Devices
Psychemedics Marijuana Screening and Confirmatory Test System (k011426)
Submission Summary (Full Text)
{0}
1
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY TEMPLATE
A. 510(k) Number:
k111929
B. Purpose for Submission:
New device
C. Measurand:
Cannabinoids
D. Type of Test:
Qualitative chemiluminescent enzyme immunoassay EIA
E. Applicant:
Psychemedics Corp.
F. Proprietary and Established Names:
Psychemedics Microplate EIA for Cannabinoids in Hair
G. Regulatory Information:
| Product Code | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| LDJ | Class II | 862.3870 – Cannabinoid test system | 91 - Toxicology |
H. Intended Use:
1. Intended use(s):
See indications for use below.
2. Indication(s) for use:
The Psychemedics Microplate EIA for Cannabinoids in Hair is an enzyme immunoassay (EIA) for the preliminary qualitative detection of cannabinoids in human head and body hair samples using a 11-nor-9-Carboxy- $\Delta$ -THC calibrator at $10~\mathrm{pg} / 10~\mathrm{mg}$ hair cutoff for the purpose of identifying marijuana use. This is an in vitro diagnostic device
{1}
intended exclusively for Psychemedics use only and is not intended for sale to anyone.
The Psychemedics Chemiluminescent Microplate EIA for Cannabinoids assay provides only a preliminary analytical test result. A more specific alternative chemical method must be used in order to obtain a confirmed analytical result. Gas Chromatography/Mass Spectrometry/Mass Spectrometry (GC/MS/MS) is the preferred confirmatory method. Clinical consideration and professional judgment should be exercised with any drug of abuse test result, particularly when the preliminary result is positive.
3. Special conditions for use statement(s):
This assay is for over the counter use.
4. Special instrument requirements:
The device is for use with a microplate reader capable of measuring at 450 and 650 nm. Plate washing also requires an instrument specifically designed to effectively and reproducibly wash all wells uniformly.
Confirmation testing is done by GC-MS/MS using a triple stage quadrupole instrument operating in the negative chemical ionization (CI) mode. The confirmatory method was cleared under k011426 and has not been modified.
I. Device Description:
The EIA screening test consists of two parts; a pre-analytical hair treatment procedure to recover the cannabinoids from the hair and the screening assay, the Psychemedics Chemiluminescent EIA for Cannabinoids in Hair. The device comprises a white microplate coated with the antigen (11-nor-9-carboxy-delta-8-tetrahydrocannabinol) conjugated to BSA, polyclonal rabbit anticannabinoid antibody, goat anti-rabbit secondary antibody conjugated to HRP (horseradish peroxidase), a chemiluminescent substrate, and plate-washing buffer. Confirmation testing is done by GC-MS/MS using a triple stage quadrupole instrument operating in the negative chemical ionization (CI) mode.
J. Substantial Equivalence Information:
1. Predicate device name(s):
Psychemedics Marijuana Screening and Confirmatory Test System
2. Predicate K number(s):
k011426
{2}
3
3. Comparison with predicate:
| Item | Device | Predicate |
| --- | --- | --- |
| Indications/Intended Use | The Psychemedics Microplate EIA for Cannabinoids in human head and body hair is an enzyme immunoassay (EIA) for the preliminary qualitative detection of cannabinoids in human head and body hair samples using a 11-nor-9-Carboxy- Δ9-THC calibrator at 10 pg/10 mg hair cutoff for the purpose of identifying marijuana use. | Same |
| Test Principle | EIA | RIA |
| Sample matrix | Hair | Same |
| Method of measurement | Microplate reader- luminescence | Gamma counter |
| Cutoff | 10 pg carboxy-THC/10 mg hair | 20 pg carboxy-THC/10 mg hair |
K. Standard/Guidance Document Referenced (if applicable):
None were referenced
L. Test Principle:
Hair extract is combined with the primary antibody in microtubes and incubated for one hour at room temperature (RT). An aliquot of the pre-incubated mixture is transferred to antigen-coated wells of the microplate. After incubation the plate is washed with wash buffer and second antibody-HRP is added. After further incubation the wells are emptied and washed with wash buffer, and chemiluminescent substrate is added. The plate is mixed, and results are read in a reader capable of reading luminescence. Results (expressed as RLU, or relative light units) are normalized by expression as B/B0 x 100.
If cannabinoids are present in the sample, less primary antibody will be bound to the solid-phase antigen, thereby resulting in less binding of HRP-labeled secondary antibody to the primary antibody; the RLU's produced by the action of the HRP on the substrate are inversely proportional to the amount of cannabinoids in the sample. Samples determined positive by the screening assay are re-weighed, washed, and by GC/MS/MS for the presence of carboxy-THC.
Confirmation testing of samples presumptively positive from the device are confirmed by first weighing out a new portion of the sample (approximately 12 mg). The samples are washed for 15 minutes at 37 °C with dry isopropanol, then three times for 30 minutes in phosphate buffer (0.01 M, pH 6.0) containing 0.1% albumin.
{3}
Samples are then dissolved and extracted for confirmation of 11-nor-9-Carboxy- $\Delta$ 9-THC by GC/MS/MS.
Carboxy-THC standard solutions are purchased from multiple vendors, prepared in the laboratory, and validated by GC/MS/MS confirmation. In addition, controls made in-house of known cannabinoid-positive hair are also employed in the assay.
# M. Performance Characteristics (if/when applicable):
# 1. Analytical performance:
# a. Precision/Reproducibility:
Precision studies were performed by taking commercially available materials consisting of carboxy-THC in methanol, to prepare spiking solutions at the following concentrations; negative, $\pm 75\%$ , $\pm 50\%$ , $\pm 25\%$ and $100\%$ of the cutoff. The concentration of each sample was confirmed by LC/MS/MS. The carboxy-THC solutions were then used to spike 15 replicates of negative hair samples. Intra-assay precision was performed in one run and inter-assay precision was performed over 5 non-consecutive days. The results are presented in the tables below:
Intra-assay
| THC (pg/10 mg hair) | Percent of Cut-off | Replicate Number | Pos/Neg |
| --- | --- | --- | --- |
| 0 | -100% | 15 | 0/15 |
| 2.5 | -75% | 15 | 0/15 |
| 5.0 | -50% | 15 | 0/15 |
| 7.5 | -25% | 15 | 0/15 |
| 12.5 | +25% | 15 | 15/0 |
| 15.0 | +50% | 15 | 15/0 |
| 17.5 | +75% | 15 | 15/0 |
| 20 | +100% | 15 | 15/0 |
{4}
Inter-assay
| THC (pg/10 mg hair) | Percent of Cut-off | Replicate Number | Pos/Neg |
| --- | --- | --- | --- |
| 0 | -100% | 75 | 0/75 |
| 2.5 | -75% | 75 | 0/75 |
| 5.0 | -50% | 75 | 0/75 |
| 7.5 | -25% | 75 | 0/75 |
| 12.5 | +25% | 75 | 75/0 |
| 15.0 | +50% | 75 | 75/0 |
| 17.5 | +75% | 75 | 75/0 |
| 20 | +100% | 75 | 75/0 |
b. Linearity/assay reportable range:
Not applicable. This is a qualitative assay.
c. Traceability, Stability, Expected values (controls, calibrators, or methods):
Psychemedics manufactures calibrators and control materials using drug stocks purchased from a commercial vendor. Each lot of drug is received with its specific certificate of analysis. The commercially obtained stock is made into the calibrators and controls to the desired concentrations. The concentrations are confirmed by GC/MS.
Stability studies for both controls and calibrators have been conducted. Protocols and acceptance criteria were described and found to be acceptable. The manufacturer claims the following expiration date for both controls and calibrators:
When stored at less than or equal to $-20^{\circ}\mathrm{C}$ for calibrators and $-10^{\circ}\mathrm{C}$ quality control the product is stable for 12 months.
d. Detection limit:
Not required since this is a qualitative test
e. Analytical specificity:
Cross-reactivity was established by spiking various concentrations of each substance into drug-free sample and evaluating the result against the cutoff control.
{5}
Results are expressed as a concentration of compound required to produce a response approximately equivalent to the cutoff concentration of the assay. The results are presented in the table below:
| Compound | Approximate concentration of compound (pg/mg) Equivalent to 10 pg/10 mg hair THC Cutoff | % Cross reactivity |
| --- | --- | --- |
| (+/-) -11-Hydroxy-Δ9-THC | 30 | 33 |
| (-) - Δ9-THC | 800 | 1.25 |
| (+/-) - Δ8-THC | 1500 | 0.67 |
| (-) - Δ8-THC | 950 | 1.05 |
| (-) -11-nor-9-Carboxy-Δ9-THC | 90 | 11 |
| (+/-) -11-nor-9-Carboxy-Δ9-THC | 350 | 2.9 |
| (-)-9-Carboxy-11-nor- Δ9-THC glucuronide | 1100 | 0.9 |
| Cannabinol | 16000 | 0.06 |
| Cannabidiol | 3600 | 0.27 |
| Nabilone | 120,000 | < 0.1 |
Structurally unrelated:
Negative hair samples were spiked with cannabinoids to -50%, and +50% of the cutoff. Structurally related and unrelated compounds were added to methanol to a concentration of 100 ng/10 mg hair then added to the negative hair sample. The following compounds do not cause interference at +/- 50% of the cutoff; acetaminophen, caffeine, dyphylline, methaqualone, theophylline, amphetamine, imipramine, methamphetamine, phencyclidine, phenmetrazine, phenylpropanolamine, amitryptiline, dextromethorphan, lidocaine, methocarbamol, nordoxepin, pentazocine, phenylephrine, triamterene, bromazepam carbamazepine, diazepam, nordiazepam, oxazepam, flurazepam, lorazepam, medazepam, temazepam, glutethimide, meprobromate, methyprylon, amitriptylinr, desipramine, doxepin, imipramine, nordoxepin, nortriptyline, protriptyline, trimipramine, butabarbital, amobarbital, secobarbital, hexobarbital, phenobarbital, ethosuximide, _-methyl---propylsuccimide, methabrital, barbital, methsuccimide, phensuccimide, N-normethsuccimide, mephenytoin, ethotoin, mephobarabital, PEMA, phenobarbital, methyl PEMA, 10, 11-dihydrocarbamazepine, primidone, carbamazepine, 5,5-diphenylhydantoin, 4-methylprimidone, tyraine, quinacrine, propanolol
There is the possibility that other substances and/or factors not listed above may interfere with the test and cause false results.
{6}
7
# Cosmetic Treatment
Tests were performed to determine the effects of various hair treatments (i.e. bleaching, dyeing, relaxer, shampoo, permanent) on samples tested using the Psychemedics Microplate EIA for THC. The ethnic origin, hair color and curvature were documented.
## Effects on Positive Samples:
60 specimens determined to be positive for THC were used in the study. The study was conducted with two different hair treatments for each hair sample. ELSIA absorbance readings before and after treatment were compared. Average changes in absorbance values after were 1.7% for bleach, -3.5% for dye, -1.75% for perm, -1.7% for relaxer and 3.8% for shampoo, where a negative sign indicates a sample becoming “more negative” due to treatment and a positive sign indicates a sample becoming “more positive.” None of the positive samples became negative after treatment.
## Effect on Negative Samples:
One hundred specimens previously determined to be negative were used in the study. The study was conducted with two different hair treatments for each hair sample. ELSIA absorbance readings before and after treatment were compared. All samples determined to be negative prior to treatment remained negative post treatment.
# Environmental Study
Twelve hair samples were contaminated with marijuana smoke and stored for two days. These samples were then split into two additional samples and soaked in either saline or water. The washed and unwashed samples were then analyzed by the screening method and confirmation method for the THC and carboxy-THC. All samples were positive by the screening method. The confirmation tests, which queries carboxy-THC, were negative for all samples.
Preliminary positive hair sample results by the screening method could be due to environmental contamination. All positive should be sent for confirmation testing on a reference method to distinguish between true positive and those samples that were positive due to external exposure.
## f. Assay cut-off:
Analytical performance of the device around the claimed cutoff is described in
{7}
precision section M.1a above
# 2. Comparison studies:
a. Method comparison with predicate device:
The Cannabinoids EIA assay is the screening component in the test system which also includes, as the second component of the cleared system (cleared under k011426), a GC/MS/MS confirmation test of carboxy-THC in hair. The study was performed by comparing EIA results against the full test system results on the same head or body hair samples. A total of 315 donor hair samples were tested first using the screening method and then confirmed by GC/MS/MS for the presence of carboxy-THC. 183 presumptively positive samples were confirmed positive by GC/MS/MS for carboxy-THC. Five presumptively positive samples were negative by GC/MS/MS for carboxy-THC with GC/MS/MS results between $50\%$ below the cutoff and the cutoff concentration. The discrepant results between screening and confirmation parts of the system are presented in the table below:
| Screening Cutoff (pg/10 mg hair) | ELSIA THC Test Results (POS/NEG) | Carboxy-THC GC/MS/MS Drug Result (pg/10 mg hair) |
| --- | --- | --- |
| 10 | POS | 0.6 (NEG) |
| 10 | POS | 0.8 (NEG) |
| 10 | POS | 0.6 (NEG) |
| 10 | POS | 0.5 (NEG) |
| 10 | POS | 0.9 (NEG) |
b. Matrix comparison:
Not applicable. The assay is intended for only one sample matrix.
# 3. Clinical studies:
a. Clinical Sensitivity:
Not applicable.
b. Clinical specificity:
Not applicable.
c. Other clinical supportive data (when a. and b. are not applicable):
Not applicable.
{8}
4. Clinical cut-off:
Not applicable.
5. Expected values/Reference range:
Not applicable.
N. Proposed Labeling:
They have not provided adequate labeling.
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.
O. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
9
Predicate graph will load when search results are available.
Embedding visualization will load when search results are available.
PDF viewer will load when search results are available.
Loading panels...
Select an item from Submissions
Click any panel, subpart, regulation, product code, or device to see details here.
Section Matches
Results will appear here.
Product Code Matches
Results will appear here.
Special Control Matches
Results will appear here.
Loading collections...
Loading
My Alerts
You will receive email notifications based on the filters and frequency you set for each alert.
Sort by:
Create Alert
Search Filters
Agent Token
Create a read-only bearer token for Claude, ChatGPT, or other agents that can call HTTP APIs.
Copy this now. It will not be shown again.
Connected apps
Apps you authorized through browser sign-in. Disconnecting revokes their access immediately.
Learn the FDA Browser
Two short videos show you everything — or skip straight to the written tutorial if you'd rather read. You can reopen this any time from the Tutorial button in the top bar.
Part 1 — Search, results, and everyday workflows 16 min
Part 2 — Embeddings: the galaxy map 3 min
1. Search: exact and fuzzy
Type a phrase like "coronary artery calcification" into the search box. You get two kinds of results. Exact results match the literal phrase — prefix searches work ("coronary artery calcificati") but suffix searches do not. Fuzzy results match on the meaning and intent of your phrase rather than the exact words, and are sorted by relevance score. Hover over the Exact or Fuzzy badge on any row to see exactly why it matched.
Use the checkboxes above the results to narrow: SaMD keeps only software-only devices, AI / ML keeps only devices with AI.
Exact vs. fuzzy search: what's the difference?
Exact matches on the literal phrase (prefix search works, suffix does not). Fuzzy matches on the meaning and intent of the phrase rather than the exact words. Hover over the badge on any row to see why it matched.
You search "coronary artery calcification" and want only software devices with AI. What two filters do you apply?
Narrow by SaMD (software-only devices), then narrow by AI/ML (devices with AI).
2. The results table
Scroll right in the results table. The intended use is extracted for you — no need to open the PDF. The device story gives a high-level snapshot of what the device does and how it's used. The AI Performance sub-table shows each output name, acceptance criteria, observed values, and development/test dataset descriptions — the same format Innolitics uses for regulatory strategy outputs, and the fastest high-level fingerprint of an AI device. It is AI-generated but has been very reliable in practice.
Where do you find a device's intended use without opening the PDF?
Scroll right in the search results table. The intended use column is extracted for you; no need to dig into the 510(k) summary PDF.
What does the AI Performance sub-table show, and why is it useful?
Output name, acceptance criteria, observed values, development dataset description, and test dataset description. It's the same format we use for regulatory strategy output and Fast 510(k) input, and the fastest high-level fingerprint of an AI device. AI-generated but reliable in practice.
3. Judging fuzzy relevance
Fuzzy results trail off in relevance as you scroll. Use three signals to decide how far down to go: the fuzzy badge explanations, the intended use column, and whether your target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, you're past the relevant zone. A top hit with a low score (~0.4) and a stretched explanation is a hint the closest predicates are far away — the project may be headed for De Novo. Note the fuzzy search is a pattern match: it doesn't handle negation ("not") well, and hardware devices can appear — filter by SaMD/AI ML to cut them.
How do you judge how far down fuzzy search results to go?
Use the relevancy signals: the fuzzy badge explanations, the intended use column, and whether the target output (e.g., Cobb angle) still appears in the AI Performance sub-table. Once it stops appearing, results are trailing off in relevancy.
4. Device detail page: chat and citations
Click a device name to open its detail page: device facts on the left, a chat window on the right. Ask something like "Describe the training data". The answer carries little citation bubbles — click one to jump to the highlighted passage in the source PDF, so you can verify every AI answer against the document. There's also a Download PDF button for sharing.
How do you verify an AI chat answer on the device detail page?
Click the citation bubbles to jump to the relevant highlight in the source document.
Reading rule for every project: how many summaries do you read in full?
At least the three most relevant 510(k) or De Novo summaries, in full. After that, use targeted chat questions to confirm your memory quickly. The tool supports this professional habit — it doesn't replace it.
5. Side-by-side comparison
Select multiple rows in the results table (aim for under ~10), then open the PDF Viewer tab. Ask one question — it goes to all selected devices in parallel, each with citations. This is the fastest way to compare and contrast devices: training data, PCCP scope, how they handled adding new scanners, and so on.
What does the side-by-side PDF viewer mode do?
Select multiple devices, open the PDF viewer tab, and ask one question (e.g., "Describe the training data"). It queries all selected devices simultaneously with citations, so you can compare and contrast quickly.
6. Collections
With rows selected, go to the Collections tab and create a labeled collection (e.g., "Cobb Angle Project"). Reload that selection any time — before a client call, pull up the collection and ask questions across all of its devices at once.
How do you save a set of selected devices for later use?
Select the rows, go to the Collections tab, and create a labeled collection (e.g., "Cobb Angle Project"). You can reload the selection anytime and carry it into the PDF viewer and other tabs that support selections.
7. Product codes and the regulations tree
Click a product code in the results to jump to it in the regulations tree — identification text, sibling product codes, and devices you can open in a PDF viewer on the right. Click a regulation number to see its identification, special controls, and related product codes. You can also search by product code or regulation number at the top of the tree. Always read the special controls if any exist for your device — it broadens your search and sharpens pre-kickoff research.
What can you do from the regulations tree view?
Browse product codes and regulation numbers, read the identification text and special controls, browse sibling product codes, open device PDFs on the right, and search by product code or regulation number at the top of the tree.
8. Chart view
Click Show Chart and segment by regulation number (or product code) to see which regulations dominate your result set. Clicking a regulation takes you into the regulations tree. Great for spotting that most matches are, say, hardware laparoscopic devices — a cue to go back and filter.
How do you see which regulations dominate a search result set?
Click "Show Chart" and segment by Regulation Number. Clicking a regulation takes you to the regulations tree.
9. The predicate graph
Open the Predicates tab for a family-tree view of predicate relationships. Click a node to trace its parents and children; selections from search carry over pre-selected. Commonly predicated devices are worth reading — a lot of people predicated them for a reason. The visual lineage is also handy on client calls, e.g. to show how a predicate family evolved and justify why your predicate still holds.
In the predicate graph, why are commonly predicated devices worth reading?
A lot of people predicated them for a reason. Clicking a node traces parents and children, and selections from search carry over pre-selected.
10. Embeddings: the galaxy map
The Embeddings tab plots every matching document in a 2-D "galaxy map" where semantically similar devices cluster together. Hover or click clusters to explore, and let AI label the clusters for you. Embeddings beat product codes for grouping: two devices can carry different product codes (LLZ vs. QIH) yet do the same thing — the embedding captures the meaning of the intended use and device story. This is also exactly how retrieval-augmented generation (RAG) works under the hood, and it makes a great visual on client calls.
Try it yourself
Head to the search page and work through a few of these AI/ML fuzzy searches to build intuition: perivascular fat on CT · aortic valve calcification opportunistic screening on noncontrast CT · breast cancer prediction on digital pathology slides · autism detection · gestational age prediction · a hearing aid that can also detect a pulse · foundation model based analysis of ECG · large language models · penetration test. Watch how the relevance scores, intended use, and AI Performance tables tell you when results stop being meaningful.