Published clinical literature; Human cadaver study; Clinical cohort study (postmenopausal women)
The literature is used to support the clinical validity of bone mineral density measurements, assess the impact of soft tissue on accuracy, and establish the clinical utility of forearm bone mineral content measurements.
Bone Densitometry; Literature Review; Clinical Accuracy; Postmenopausal Women
Clinical Evidence
Study Design
Population
Comparator
Key Endpoints
Bone Changes Occurring Spontaneously and Caused by Oestrogen in Early Postmenopausal Women; Longitudinal clinical cohort; Follow-up/Duration: 1 year; Study Period: 1986
52 early postmenopausal women; Sample Size: 52
Placebo group
Bone mineral content and density changes
Impact of Soft Tissue on In Vivo Accuracy of Bone Mineral Measurements in the Spine, Hip, and Forearm: A Human Cadaver Study; Cadaver study; Study Period: 1995
14 human cadavers; Sample Size: 14
Not applicable for this study
In vivo accuracy errors (SEE%)
Indications for Use
To estimate bone mineral content (BMC, grams) and bone mineral density (BMD, grams/cm²).
Device Story
The DTX-200 is a dual-energy X-ray absorptiometry (DXA) bone densitometer designed to measure bone mineral content (BMC) and bone mineral density (BMD) in the forearm. The device utilizes an X-ray generator (55 kV, 300 μA) with a tin (Sn) filter to produce dual-energy peaks (approx. 29 keV and 46 keV). A sandwich-construction solid-state photodiode detector captures the transmitted X-rays. The system processes the input data by dividing the forearm into two compartments—bone and soft tissue—and solving two simultaneous equations to quantify bone mass. The device is operated by clinical staff in a medical setting. Output is provided to the healthcare provider to assist in monitoring bone changes and clinical follow-up. The device benefits patients by providing a precise, non-invasive method for assessing skeletal bone mineral status.
Clinical Evidence
Evidence includes two manufacturer-sponsored studies and five literature citations. A functional correlation study (n=81) showed high correlation with the DTX-100 predicate. A precision study (n=15) confirmed the device's ability to detect clinically meaningful differences. Literature supports the clinical utility of forearm BMC measurements, the precision of X-ray absorptiometry, and the accuracy of bone mineral measurements in the presence of soft tissue.
Technological Characteristics
Dual-energy X-ray absorptiometry (DXA) system. X-ray source: 55 kV, 300 μA, Sn-filtered. Detector: Solid-state photodiode sandwich construction. Pixel resolution: 0.4mm x 0.4mm. Source collimation: 1 mm. System includes computer, monitor, and printer. Operates via line-by-line scanning.
Indications for Use
Indicated for the estimation of bone mineral content and bone mineral density in the forearm for clinical assessment of bone mass.
Regulatory Classification
Identification
A bone densitometer is a device intended for medical purposes to measure bone density and mineral content by x-ray or gamma ray transmission measurements through the bone and adjacent tissues. This generic type of device may include signal analysis and display equipment, patient and equipment supports, component parts, and accessories.
Predicate Devices
Osteometer MediTech A/S model DTX-100 bone densitometer
Hologic, Inc. model QDR-1000/W bone densitometer
Submission Summary (Full Text)
{0}
JUL - 5 1996
K961011
# Appendix G
510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
21 CFR 807.92(a):
21 CFR 807.92(a)(1):
* Submitter's name and address:
Osteometer Meditech A/S
Glerupvej 2, DK-2160 Rodovre
Denmark
* Submitter's telephone number: 011 45 44 92 42 00
* Contact person:
Ms. Anja Koue
Official Correspondent
Osteometer Meditech A/S
Glerupvej 2, DK-2160 Rodovre
Denmark
* Date this 510(k) summary was prepared: March 11, 1996
21 CFR 807.92(a)(2):
* Trade/proprietary name of the device: DTX-200 Bone Densitometer
* Classification name: Bone densitometer
21 CFR 807.92(a)(3); Legally marketed device (predicate device) to which equivalence is claimed:
* Osteometer MediTech A/S model DTX-100 bone densitometer
* Hologic, Inc. model QDR-1000/W bone densitometer
21 CFR 807.92(a)(4); Description of the device that is the subject of this premarket notification:
The DTX-200 candidate device is a dual energy system, where the measured object, in this case the forearm, by the software, is divided into two compartments, bone and soft-tissue. Having two energies, and two compartments (unknowns) it is possible to establish and solve two formulae with two unknowns.
The X-Ray generator is driven at 55 kV, with a current of 300
OM11
1 of 9
491
{1}
OM11
2 of 9
492
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
μAmps. The beam is filtered with Tin (Sn), using the K-Edge of this material to enhance the separation of the low and the high energy. The average energy peaks are approximately 29 keV, and 46 keV respectively. A highly stable X-Ray output, together with a source collimation of 1 mm and a pixel resolution of 0.4mm X 0.4mm, yields images of very high quality.
The detector is a sandwich construction, using solid state photodiodes consisting of two diodes with scintillation material. The two energies are separated by the sandwich construction. The X-ray beam meets the first scintillator, where predominantly the low energy is detected, and then it meets the second scintillator, where the rest of the energy is detected.
The typical time to estimate bone mineral content and bone mineral density is approximately 4 minutes. The effective patient dose is estimated to be 0.1 micro Sievert per scan.
## 21 CFR 807.92(a)(5); Intended use:
The intended use of the DTX-200 candidate device is the same as the labeled intended use of the predicate devices to which equivalence is claimed; i.e., "To estimate bone mineral content (BMC, grams) and bone mineral density (BMD, grams/cm²).
## 21 CFR 807.92(a)(6); Technological characteristics:
The design, material, chemical composition, energy source and other technological characteristics of the subject device are considered to be the same as the technological characteristics of the predicate devices. A summary of the technological characteristics of the subject device in comparison to those of the predicate devices follows:
* X-ray transmission source:
Both the DTX-200 subject and the DTX-100 predicate bone densitometer devices consist of a computer, keyboard, monitor, printer, and scanner system using an X-ray transmission source. The DTX-200 candidate device uses dual X-ray photon absorptiometry (DXA), the DTX-100 predicate device uses single X-ray photon absorptiometry (SXA), and the QDR-1000/W predicate device uses dual energy quantitative digital radiography to estimate bone mineral content (BMC) and bone mineral density (BMD).
* Soft-tissue equivalent material:
{2}
OM11
3 of 9
493
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
In contrast to the DTX-100 predicate device (that uses water as soft tissue equivalent material) and the QDR-1000/W predicate device (that uses a calibration wheel), the DTX-200 candidate device does not require the use of water as an soft tissue equivalent material. The precision of the devices is comparable.
### Calibration:
Both the DTX-200 candidate device and the DTX-100 predicate device use a line-by-line calibration system, ensuring good precision. The QDR-1000/W predicate device self calibrates using an automatic internal reference system.
### Imaging technique, data acquisition, and quality control phantom:
These characteristics are the same on the DTX-200 candidate device as the DTX-100 predicate device.
### Source collimation:
The source collimation for both the DTX-200 candidate device and the DTX-100 predicate device is the 1 mm diameter. The source collimation for the QDR-1000/W predicate device is pencil beam, 0.09 inch diameter.
### Reference location:
The reference location for both the DTX-200 candidate device and the DTX-100 predicate device is the 8 mm Ulna-Radius gap. The reference location for the QDR-1000/W predicate device is the ulnar styloid.
### Accuracy and precision error:
Accuracy and precision error of the DTX-200 candidate device is comparable to that of the DTX-100 predicate device and the QDR-1000/W predicate device.
21 CFR 807.92(b);
{3}
OM11
4 of 9
494
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
510(k) summaries for those premarket submissions in which a determination of substantial equivalence is also based on performance data shall also contain the following:
21 CFR 807.92(b)(1); Brief discussion of the nonclinical tests submitted, referenced, or relied on in this premarket notification submission:
There were no nonclinical tests submitted, referenced, or relied on in this submission.
21 CFR 807.92(b)(2); Brief discussion of the clinical tests submitted, referenced, or relied on in this premarket notification submission:
The results of seven (7) clinical tests were submitted, referenced, or relied on in this premarket submission. Two clinical tests were sponsored by the device manufacturer and five of the clinical studies were from published literature:
1. Two clinical tests sponsored by the device manufacturer; (A) a functional correlation study between the DTX-200 candidate device and the DTX-100 predicate device and (B) a precision study of the DTX-200 candidate device.
A. Functional Correlation Study: Bone mineral content (BMC) and bone mineral density (BMD) were measured at the 8 mm distal site and the ultra distal site in the nondominant forearm of eighty one (81) individuals to determine the functional relationship and to evaluate the degree of correlation between the DTX-200 candidate device and the DTX-100 predicate device. Linear regression analysis was used to determine the degree of correlation.
B. Precision Study: Bone mineral content (BMC) and bone mineral density (BMD) were measured at the 8 mm distal site and the ultra distal site in the nondominant forearm of fifteen (15) individuals, five (5) times each, to estimate the precision error of the DTX-200 candidate device. The precision estimates were calculated as the coefficient of variation at a 95% confidence interval using log-transformed data.
2. Five of the clinical studies were from published literature:
A. "Bone Changes Occurring Spontaneously and Caused by Oestrogen in Early Postmenopausal Women: A Local or
{4}
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
Generalized Phenomenon?" Published in the British Medical Journal, 26th. April 1986, by Anders Gotfredsen, Lisbeth Nilas, Bente Juel Riis, Karsten Thomsen, and Claus Christiansen.
### Abstract of this clinical study:
"Regional values of bone mineral content and bone mineral density were calculated from total body and dual photon absorptiometry scans of 52 early postmenopausal women treated with placebo. The six regions were head, arms, chest, spine, pelvis, and legs. In addition, bone mineral density of the lumbar spine was measured by dual photon absorptiometry and bone mineral content of the forearm using single photon absorptiometry, using separate special purpose scanners. All regions were unchanged after one year of treatment with oestrogen, excluding the lumbar spine, for which values rose. Values for all regions except the lumbar spine fell significantly in the placebo group. The rates of loss ranged from 2% to 8%, with no significant differences among the regions."
B. "Single X-Ray Absorptiometry of the Forearm: Precision, Correlation, and Reference Data". Published in Calcified Tissue International, in 1994, by T.L. Kelly, G. Crane, and D.T. Baran.
### Abstract of this clinical study:
"The performance of a single X-ray absorptiometry (SXA) device incorporating an X-ray tube as a photon source was evaluated with respect to precision in vivo and in vitro, scan time, image quality, and correlation with an existing dual energy X-ray absorptiometry (DXA) device. SXA precision in vivo, expressed as a coefficient of variation (CV), was 0.66% for bone mineral content (BMC) and 1.05% for bone mineral density (BMD). Precision in vitro, based on 78 measurements of a forearm phantom over 195 days, was 0.53%. Correlation with DXA at the 8 mm distal forearm site was high (r = 0.97 for BMC and r = 0.96 for BMD). A preliminary SXA reference data base composed of 151 healthy Caucasian American women was developed to facilitate the interpretation of patient measurements. SXA scan time was 4 minutes
OM11
5 of 9
{5}
# Appendix G
510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
and delivered a radiation exposure of 1.68 mrem. SXA image quality and spatial resolution were superior to SPA and comparable to DXA."
C. "Impact of Soft Tissue on In Vivo Accuracy of Bone Mineral Measurements in the Spine, Hip, and Forearm: A Human Cadaver Study". Published the Journal of Bone and Mineral Research, Volume 10, Number 6, 1995, by Ole L. Svendsen, Christian Hassager, Vera Skodt, and Claus Christiansen.
Abstract of this clinical study:
"The impact of soft tissue in vivo based on bone mineral density (BMD) measurements of the spine and hip by dual energy X-ray absorptiometry and of the forearm by single photon absorptiometry was assessed by use of 14 human cadavers. The in vivo accuracy errors (SEE%) were: forearm 3-5%, anteroposterior spine 5.3%, lateral spine 10-12%, and femoral greater trochanter, neck, total, intertrochanteric, and Ward's triangle 3%, 6.5%, 6.7%, 8%, and 11-13%, respectively. Except from the lateral spine and the greater trochanter, the slopes of the linear regressions of in vivo BMD against an in vitro BMD were not significantly different from 1(p > 0.05). The calculated random accuracy error of BMD measurements due to fat inhomogeneity was estimated to 3-4% for the anteroposterior spine and 9-14% for the lateral spine (from abdominal computed tomography in 26 healthy women)."
D. "Dual-Energy X-Ray Absorptiometry: A Precise Method of Measuring Bone Mineral Density in the Lumbar Spine". Published in The Journal of Nuclear Medicine, July 1990, by Marc A. Hansen, Christian Hassager, Kirsten Overgaard, Ulla Marslew, Bente J. Riis, and Claus Christiansen
Abstract of this clinical study:
"We compared two methods of measuring spinal bone mineral content and density (BMC/BMD): conventional dual-photon absorptiometry (DPA) and a more recent method, dual energy X-ray absorptiometry (DEXA). The clinical usefulness of both methods was compared in the measurement of BMC in the forearm. DEXA had a long term in vivo precision of 1% which
OM11
6 of 9
{6}
OM11
7 of 9
497
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
was significantly better than that of DPA. Changes in the distribution of fatty tissue influenced the accuracy of the two spinal methods in different ways. Forearm BMC discriminated between the bone mass of early and late postmenopausal women to the same degree as DPA and DEXA. The variability in the response to the estrogen treatment and placebo with DEXA and forearm BMC than with DPA."
E. "Patient Dose in Dual X-Ray Absorptiometry". Published in Osteoporosis International, (1994) 4:11-15, by M.K. Lewis, G.M. Blake, and I. Fogelman.
Abstract of this clinical study:
"Dual X-ray absorptiometry (DXA) provides a convenient, non-invasive method of assessing skeletal bone mineral which is widely used for clinical studies. This report describes a study to estimate dose of radiation [ICRP-60 (1990)] to a female patient from scans performed on three DXA scanners: the Hologic QDR-1000, QDR-1000/W and QDR-2000. The scans modes studied were: total body; anteroposterior (AP) lumbar spine; lateral lumbar spine; proximal femur; distal forearm. An ion chamber and tissue-equivalent phantom were used to determine entrance surface dose and percentage depth-dose curves for each scan mode. Anatomical data from ICRP-23 (Reference Man) and a body section atlas were used to estimate the absorbed dose to each organ in the scan fields. Effective dose was estimated using the ICRP-60 tissue weighing factors and the fraction of each organ in the scan field.
21 CFR 807.92(b)(3; The conclusions drawn from the nonclinical and clinical tests that demonstrate that the subject device is as safe, as effective, and performs as well as or better than the predicate device:
Clinical studies sponsored by Osteometer MediTech A/S:
1A. Functional Correlation Study: The clinical study on eighty one (81) patients showed the DTX-200 candidate device and the DTX-100 predicate device are substantially equivalent in their ability estimate bone mineral content (BMC) and bone
{7}
OM11
8 of 9
498
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
mineral density (BMD).
1B. Precision Study: The results of the clinical study on fifteen (15) individuals shows that the DTX-200 candidate device estimates bone mass with precision errors in an order that makes it possible to detect clinically meaningful differences between repeated bone scans in the same individual.
Clinical studies published in the literature:
2A. "Bone Changes Occurring Spontaneously and Caused by Oestrogen in Early Postmenopausal Women: A Local or Generalized Phenomenon?" This study concluded, "Loss of bone in early menopause is a generalized phenomenon, affecting all parts of the skeleton. Furthermore, oestrogen prophylaxis for loss of bone is effective in all parts of the skeleton. Finally, it is suggested that the measurement of bone mineral content in the forearm should be used for clinical follow up of bone changes, as this method is superior to others in the ratio of change to precision."
2B. "Single X-Ray Absorptiometry of the Forearm: Precision, Correlation, and Reference Data": This study concluded, "SXA image quality and spatial resolution were superior to SPA and comparable to DXA."
2C. "Impact of Soft Tissue on In Vivo Accuracy of Bone Mineral Measurements in the Spine, Hip, and Forearm: A Human Cadaver Study": This study concluded, "Acceptable accuracy errors below 6-7% (of soft tissue in vivo) of BMD measurements were obtained in the anteroposterior spine, the forearm, and the neck, greater trochanter, and total proximal femur."
2D: "Dual-Energy X-Ray Absorptiometry: A Precise Method of Measuring Bone Mineral Density in the Lumbar Spine": This study concluded, "that DEXA provides a fast and precise measurement of spinal BMC/BMD. The accuracy remains to be evaluated for in vivo studies"
2E. "Patient Dose in Dual X-Ray Absorptiometry": This study concluded, "PDD (percentage depth dose) curves for the QDR-1000 and the QDR-1000/W part-body scans were found to be identical, with ESD (entrance skin dose) varying according to the speed of the scanning arm. The QDR-2000 pencil beam PDD curves were identical to those for the QDR-1000 and the QDR-1000/W with the ESD a factor of 1.6 higher due to the use of a lighter scanning table. The PDD curves for the QDR-2000 1-
{8}
# Appendix G
## 510(k) Summary
[This 510(k) summary is furnished in accordance with 21 CFR 807.92]
min, 2-min and 3-min fan beam spine and hip scans were identical with ESD measurements varying in proportion with scanning speed. As would be expected from the ESD and PDD curve data the effective doses for scans performed on the QDR-1000 and QDR-1000/W were identical while those for the QDR-2000 pencil beam mode were higher by a factor of 1.6."
... END OF 510(k) SUMMARY ...
OM11
9 of 9
499
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.