K120813 · Andon Medical Co., Ltd. · NBW · Dec 7, 2012 · Clinical Chemistry
Device Facts
Record ID
K120813
Device Name
BLOOD GLUCOSE MONITORING SYSTEM
Applicant
Andon Medical Co., Ltd.
Product Code
NBW · Clinical Chemistry
Decision Date
Dec 7, 2012
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 862.1345
Device Class
Class 2
Attributes
Software as a Medical Device, Pediatric
Indications for Use
iHealth BG3 Smart Gluco-Monitoring System is intended to be used for: · quantitative measurement of glucose in fresh capillary whole blood samples drawn from the fingertip, palm, forearm, upper arm, calf or thigh · single person measurement only and should not be shared · self-testing outside the body (in vitro diagnostic use) by people with diabetes at home as an aid to monitor the effectiveness of diabetes control The iHealth BG3 Smart Gluco-Monitoring System should not be used for the diagnosis of or screening for diabetes, or for neonatal use. Alternative Site Testing (AST) should be done only during steady state times when glucose levels are not changing rapidly. The AGS10001 test strips are intended for use with the iHealth BG3 meter to quantitatively measure glucose in fresh capillary whole blood samples drawn from the fingertips, palm, forearm, upper arm, calf or thigh using the iHealthBG3 meter
Device Story
iHealth BG3 Smart Gluco-Monitoring System is an electrochemical biosensor for capillary whole blood glucose monitoring. Device consists of a meter, test strips, lancets, and control solutions. User applies blood to test strip; meter uses capillary action and amperometry to measure glucose; results are transmitted to an iPhone or iPod touch via a physical dock connection. The iOS device provides the user interface, display, and data storage. Intended for home use by patients with diabetes to monitor glucose levels. System supports up to 10,000 records with date/time stamps. Benefits include integration with mobile platforms for easier data management and tracking of diabetes control. Healthcare providers use the displayed results to assess treatment effectiveness. Device is for single-patient use only.
Clinical Evidence
Conforms to ISO 15197 standards for in vitro diagnostic blood-glucose monitoring systems. Performance validated through non-clinical and clinical testing. No specific numerical clinical trial results (e.g., sensitivity/specificity) provided in the summary document.
Technological Characteristics
Electrochemical biosensor; glucose oxidase enzyme; amperometry detection. Dimensions: 102mm x 58mm x 22mm. Power: 3.3V DC (via connected iOS device). Connectivity: Docking interface for iPhone/iPod touch. Standards: ISO 15197, IEC 61010-1, EN 61326. Single-patient use; validated for disinfection with CaviWipes.
Indications for Use
Indicated for people with diabetes for quantitative self-testing of capillary whole blood glucose from fingertip, palm, forearm, upper arm, calf, or thigh. Contraindicated for neonates, diagnosis/screening of diabetes, and patients who are dehydrated, hypotensive, in shock, in hyperglycemic-hyperosmolar state, or critically ill.
Regulatory Classification
Identification
A glucose test system is a device intended to measure glucose quantitatively in blood and other body fluids. Glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, and idiopathic hypoglycemia, and of pancreatic islet cell carcinoma.
Special Controls
*Classification.* Class II (special controls). The device, when it is solely intended for use as a drink to test glucose tolerance, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.
Predicate Devices
AG-608N Single Blood Glucose Monitoring System (K110017)
Submission Summary (Full Text)
{0}
510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION
DECISION SUMMARY
ASSAY AND INSTRUMENT COMBINATION TEMPLATE
A. 510(k) Number:
k120813
B. Purpose for Submission:
Modification to cleared meter to connect and provide User Interface with iOS iPhone and iPod
C. Measurand:
Capillary Whole Blood Glucose
D. Type of Test:
Quantitative, electrochemical biosensor, glucose oxidase
E. Applicant:
ANDON MEDICAL CO., LTD
F. Proprietary and Established Names:
iHealth BG3 Smart Gluco-Monitoring System
G. Regulatory Information:
1. Regulation section:
21CFR Sec.-862.1345 Glucose test system.
21CFR Sec.-862.1660 Quality control material (assayed and unassayed).
21CFR Sec.-862.2100 Calculator/data processing module for clinical use.
2. Product code:
NBW - system, test, blood glucose, over the counter
CGA - Glucose Oxidase
JQP - calculator/data processing module, for clinical use
3. Classification:
Class 2
Class 1, subject to limitations 21CFR862.9(c)(5) respectively
{1}
4. Panel:
Clinical Chemistry (75)
H. Intended Use:
1. Intended use(s):
See indications for use below
2. Indication(s) for use:
iHealth BG3 Smart Gluco-Monitoring System is intended to be used for:
- quantitative measurement of glucose in fresh capillary whole blood samples drawn from the fingertip, palm, forearm, upper arm, calf or thigh
- single person measurement only and should not be shared
- self-testing outside the body (in vitro diagnostic use) by people with diabetes at home as an aid to monitor the effectiveness of diabetes control
The iHealth BG3 Smart Gluco-Monitoring System should not be used for the diagnosis of or screening for diabetes, or for neonatal use.
Alternative Site Testing (AST) should be done only during steady state times when glucose levels are not changing rapidly.
The AGS1000I test strips are intended for use with the iHealth BG3 meter to quantitatively measure glucose in fresh capillary whole blood samples drawn from the fingertips, palm, forearm, upper arm, calf or thigh using the iHealthBG3 meter.
3. Special conditions for use statement(s):
- For in vitro diagnostic use only
- For over-the-counter use
- For single-patient use only
- Not intended for use on neonates
- Not for the diagnosis of or screening for diabetes mellitus
- Not for use on patients who are dehydrated, hypotensive, in shock, or for individuals in hyperglycemic-hyperosmolar state, with or without ketosis.
- Not for use in critically ill patients
- AST results should not be used to calibrate a continuous glucose monitor (CGM) or in insulin dose calculations
2 of 12
{2}
4. Special instrument requirements
iHealth BG3 Meter
I. Device Description:
iHealth BG3 Smart Gluco-Monitoring System consist of blood glucose meter, test strips, sterile lancets, lancing device and AGS1000I Control Solutions (Level I, Level II and Level III) cleared under k110017.
The device iHealth BG3 Smart Gluco-Monitoring System is based on an electrochemical biosensor technology (electrochemical) and the principle of capillary action. Capillary action at the end of the test strip draws the blood into the action chamber and the blood glucose result is displayed in 5 seconds. The control solution, is used to test the performance of the device.
The appearance of the subject iHealth BG3 Smart Gluco-Monitoring System is different from the predicate device. The subject device iHealth BG3 Smart Gluco-Monitoring System must connect to an iPhone or iPod touch to display the results and complete its function.
J. Substantial Equivalence Information:
1. Predicate device name(s):
AG-608N Single Blood Glucose Monitoring System
2. Predicate 510(k) number(s):
K110017
3 of 12
{3}
3. Comparison with predicate:
| CHARACTERISTICS | SUBJECT DEVICE: iHealth BG3 Smart Gluco-Monitoring System | PREDICATE: AG-608N Single Blood Glucose Monitoring System (K110017) |
| --- | --- | --- |
| Indications for use | iHealth BG3 Smart Gluco-Monitoring System is intended to be used for: • quantitative measurement of glucose in fresh capillary whole blood samples drawn from the fingertip, palm, forearm, upper arm, calf or thigh • single person measurement only and should not be shared • self-testing outside the body (in vitro diagnostic use) by people with diabetes at home as an aid to monitor the effectiveness of diabetes control The iHealth BG3 Smart Gluco-Monitoring System should not be used for the diagnosis of or screening for diabetes, or for neonatal use. Alternative Site Testing (AST) should be done only during steady state times when glucose levels are not changing rapidly. The AGS1000I test strips are intended for use with the iHealth BG3 meter to quantitatively measure glucose in fresh capillary whole blood samples drawn from the fingertips, palm, forearm, upper arm, calf or thigh using the iHealthBG3 meter. | Same |
{4}
| Detection Method | Amperometry | Same |
| --- | --- | --- |
| Enzyme | Glucose Oxidase | Same |
| Type of Meter | Biosensor (Electrode) | Same |
| Sample Source | Capillary whole blood from finger, palm, forearm, upper arm, calf and thigh | Same |
| Sample Application | Blood sample is placed directly on the test strip after finger or AST is lanced. | Same |
| Hematocrit Range | 20-60% | Same |
| Operating Temperature Range | 10°C~35°C (50°-95°F) | 10°C ~40°C (50°-104°F) |
| Dimensions | 102mm×58mm ×22mm | 52mm x 92mmx |
| Display | Connect to iPhone or iPod touch to display measurement results | LCD |
| Result Presentation | mg/dL or mmol/L | Same |
| Memory Capabilities | 10000 times with time and date displaying | 500 times with time and |
| Test Start | Automatic | Same |
| Test Time | 5 second | Same |
| Power Source | DC 3.3V (Powered by iOS device connected to the meter) | DC 3V (CR2032) |
| Battery Life | N/A | Approx. 500 |
| Measurement Range | 20mg/dL-600mg/dL (1.1mmol/L~33.3mmol/L) | Same |
| Qualified Test Strip | AGS-1000I Test Strip | AGS-1000N Test |
| Sample Volume | Minimum 0.7 micro liter | Same |
{5}
6 of 12
| Other function | N/A | USB function.
Voice function |
| --- | --- | --- |
K. Standard/Guidance Document Referenced (if applicable):
ISO 15197, In vitro diagnostic test systems - Requirements for in vitro whole blood glucose monitoring systems intended for use by patients for self-testing in management of diabetes mellitus, First Edition 2003-05-01, Approved. (InVitro Diagnostics)
IEC 61010-1, Safety requirements for electrical equipment for measurement, control, and laboratory use, 2001
EN 61326:2006 Electrical equipment for measurement, control and laboratory use
L. Test Principle:
The test is based on electrochemical biosensor technology and the principle of capillary action. The electrical current generated by the reaction of glucose with the reagent of the strip is measured by the meter and is displayed on an iPhone or iPod touch as the corresponding blood glucose level. The strength of the current produced by the reaction depends on the amount of glucose in the blood sample.
M. Performance Characteristics (if/when applicable):
The device is a modification of a cleared AG-608N meter (k110017) to connect and provide the User Interface with iOS iPhone and iPod using the same test strip and control cleared with the AG-608N meter in k110017. Meter, strip and analytical module are unchanged. The sponsor conducted necessary validation of the device, such as human factors testing and bench testing (see Section P).
1. Analytical performance:
a. Precision/Reproducibility:
As established in k110017
b. Linearity/assay reportable range:
As established in k110017
c. Traceability, Stability, Expected values (controls, calibrators, or methods):
As established in k110017
{6}
d. Detection limit:
As established in k110017
e. Analytical specificity:
As established in k110017
f. Assay cut-off:
As established in k110017
2. Comparison studies:
a. Method comparison with predicate device:
As established in k110017
b. Matrix comparison:
As established in k110017
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):
The sponsor performed a consumer study in k110017 to test the effects of 100 lay-users on the performance of their candidate device. The results of this testing are described in k110017.
The performance established in k110017 was confirmed with Consumer Accuracy data derived from the Human Factors (HF) study described in section P of this document. The results* using 30 meters and iPhones are described below:
System Accuracy Results comparing meter finger results to YSI
Glucose concentrations < 75 mg/dL
7 of 12
{7}
| Within ±5 mg/dL | Within ±10 mg/dL | Within ±15 mg/dL |
| --- | --- | --- |
| 4/6 | 6/6 | 6/6 |
| 67% | 100% | 100% |
Glucose concentrations $\geq 75\mathrm{mg / dL}$
| Within ±5% | Within ±10% | Within ±15% | Within ±20% |
| --- | --- | --- | --- |
| 10/24 | 19/24 | 23/24 | 24/24 |
| 42% | 79% | 96% | 100% |
* Note: HF studies were conducted on 60 systems (30 iPhones and 30 iPods), however, only the 30 iPhones were evaluated for data accuracy. The data tables above reflect the performance data generated from the 30 iPhones.
4. Clinical cut-off: Not Applicable
5. Expected values/Reference range:
| Time of day | People without diabetes |
| --- | --- |
| Fasting and before meals | <100 mg/dL |
| 2 hours after meals | <140 mg/dL |
(1) American Diabetes Association: Diagnosis and Classification of Diabetes Mellitus (Position Statement). Diabetes Care 34 (Supp. 1) S66, 2011.
(2) Tietz Fundamentals of Clinical Chemistry, 6th Edition, Edited by Burtis CA and Ashwood ED, W. B. Saunders Co., Philadelphia, PA, 2008, p. 849.
N. Instrument Name:
iHealth BG3 Meter
O. System Descriptions:
1. Modes of Operation:
{8}
Does the applicant's device contain the ability to transmit data to a computer, webserver, or mobile device:
Yes ☐ or No ☑
Does the applicant's device transmit data to a computer, webserver, or mobile device using wireless transmission?:
Yes ☑ or No ☐
2. Software:
FDA has reviewed applicant's Hazard Analysis and software development processes for this line of product types:
Yes ☑ or No ☐
3. Specimen Identification:
There is no sample identification function with this device. Samples are applied directly to the test strip as they are collected.
4. Specimen Sampling and Handling:
This device is intended to be used with capillary whole blood from the finger and other alternative sites (palm, forearm, upper-arm, calf and thigh). Since the whole blood sample is applied directly to the test strip, there are no special handling or storage issues.
5. Calibration:
Auto-coding: No calibration is required from the user, since all test strips have the same one-code.
6. Quality Control:
No quality control (QC) tracking software is provided. The user is advised to monitor QC material when testing is within labeled range. Three levels of
9 of 12
{9}
aqueous glucose control solutions are available with this system. Control solution Level 2 is provided with the kit. Recommendations on when to test the control materials are provided in the labeling. The control solution readings are not included in the average of the patient results. An acceptable range for each control level is printed on the test strip vial label. The user is cautioned not to use the meter if the control result falls outside these ranges.
## P. Other Supportive Instrument Performance Characteristics Data Not Covered In The "Performance Characteristics" Section above:
Applicable temperature and humidity studies; sample volume study; Altitude study; hematocrit study were established in k110017
1. A readability assessment was conducted by SMOG method on the instructions for use, and they were found to be at 8th grade reading level or below.
2. Human Factors testing
A human factors study was conducted with 30 lay user participants, using 30 iPhones, 30 iPod Touch devices, 60 iHealth BG3 meters (30 will dock iPhone and 30 will dock iPod Touch) and 3 lots of test strips. As part of the human factors study, the 30 lay user participants used their own fingersticks to evaluate data accuracy to the 30 iPhones, as described in section M.3.c. above. The study participants had varied, but not advanced computer skills. The first 10 users used test strip lot 1, the second 10 users used test strip lot 2, and the last 10 users used test strip lot 3. The users were given labeling in English and instructed to set up and use the system as intended to be used. Results of this study are described in section M.3.c above.
Lay user questionnaire: Lay users evaluated the ease of use of the device and the presentation of the labeling. All users thought that the iHealth BG3 Smart Gluco-Monitoring System was easy to use and most answered "easy" and "very easy" to each question.
3. Bench testing performed for data from meter to iPhone or iPod device
Bench testing was performed to verify the transmission accuracy of electrical signals between Apple's mobile operating system for iPhone and iPod platforms (iOS) and glucose meter. This evaluation consisted of the following: sending of data (output from the glucose meter) and receiving of data (input
10 of 12
{10}
into the iOS device). Data from the glucose meter and the receiving data on iOS device, was in 100% agreement
4. Bench testing memory limits and rollover synchronization.
Testing was performed to evaluate that the iOS and device can save 10000 records and that the records can rollover properly in the iOS app. The criteria for successful rollover of data were that when the memory is full, the new record will replace the record that existed from the first record. The stated acceptance criteria was met for this study
5. Electromagnetic Compatibility (EMC) testing was performed/ passed. A certificate verifying that EMC testing was performed and passed was provided.
6. The applicant provided software documentation and testing that supports the device was developed and is under good software lifecycle processed.
7. Disinfection studies/Infection Control Studies: The devices are intended for single-patient use (iHealth BG3 Smart Gluco-Monitoring System). Disinfection efficacy studies were performed on the materials comprising the meters and lancing device by outside commercial testing demonstrating complete inactivation of hepatitis B virus (HBV) with the chosen disinfectant, CaviWipes (EPA Registration #46781-8). Robustness studies were also performed by the sponsor demonstrating that there was no change in performance or external materials for each of the meters and lancing device after 11,000 cleanings and 11,000 disinfection steps with the CaviWipes. The robustness studies were designed to stimulate 5 years of single-patient use. Labeling was reviewed for adequate instructions for the validated cleaning and disinfection procedures.
The sponsor provides a disposable sleeve for the iPod or iPhone to be used with each test strip, not to be reused and to be disposed of as bio-hazard, therefore no disinfection studies were required of the iPod or iPhone.
8. Customer service is available Monday through Friday 8:30 am to 5:30 pm Pacific Standard Time at the following toll-free number: 1-855-816-7705. The labeling instructs patients to contact their healthcare provider at all other times for help)
11 of 12
{11}
Q. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.
R. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
12 of 12
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.