REGIUS CONSOLE CS-2000 AND REGIUS CONSOLE CS-3000, MEDICAL IMAGE PROCESSING WORKSTATION
K051523 · Konica Minolta Medical & Graphic, Inc. · LLZ · Jul 20, 2005 · Radiology
Device Facts
Record ID
K051523
Device Name
REGIUS CONSOLE CS-2000 AND REGIUS CONSOLE CS-3000, MEDICAL IMAGE PROCESSING WORKSTATION
Applicant
Konica Minolta Medical & Graphic, Inc.
Product Code
LLZ · Radiology
Decision Date
Jul 20, 2005
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 892.2050
Device Class
Class 2
Indications for Use
Receive and process electronic images of patients. The REGIUS CS-2000/CS-3000 is NOT intended for use with digital mammography system.
Device Story
Medical image processing workstation; receives electronic images from connected CR readers (direct or cassette type); processes images for display and output to printers or host computers. Operated by radiographers in clinical settings; provides tools for contrast adjustment, frequency enhancement (F, E, H-processing), masking, rotation, flipping, re-sampling, resizing, stitching, and grid suppression. CS-3000 includes hard disk storage; CS-2000 uses DCT compression for floppy disk output. Optional portable mode uses PDA for barcode registration of cassettes in wards. Output displayed on touch-panel console; assists radiographers in image preparation; not intended for diagnostic purposes.
Clinical Evidence
Bench testing only. No clinical data provided. Device safety and efficacy supported by hazard analysis and compliance with software guidance documents.
Technological Characteristics
Off-the-shelf PC hardware (Pentium 4, 512MB-1GB RAM, 80GB HDD); 15" 1MP touch-screen LCD; Ethernet (IEEE 802.3) and IEEE 1394 connectivity. Software runs on Windows XP. Image processing includes frequency enhancement, grid suppression, and stitching. Complies with UL 60950, IEC/CISPR 22, and IEC/CISPR 24.
Indications for Use
Indicated for receiving and processing electronic patient images. Not intended for use with digital mammography systems.
Regulatory Classification
Identification
A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.
Special Controls
*Classification.* Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).
{0}------------------------------------------------
K051523
510(k) Summary
as required by 807.92
- 1. Company Identification
## JUL 2 0 2005
KONICA MINOLTA MEDICAL & GRAPHIC, INC. 2970 Ishikawa machi Hachioji-shi, Tokyo 192-8505, Japan Tel: +81-426-60-9607 Fax: +81-426-60-9588
- 2. Official Correspondent
Koji Kubo(Mr.) Department TS Advanced Technology Division R & D Center
- 3. Date of Submission
June 6, 2005
- 4. Device Trade name, Common Name
REGIUS CONSOLE CS-2000 and REGIUS CONSOLE CS-3000, Medical Image Processing Workstation
- 5. Classification
Class II, 90-LLZ, 21CFR 892.2050, Picture archiving and communications system
- 6. Intended Use
Receive and process electronic images of patients
- 7. Applicable mandatory and voluntary standards
REGIUS CONSOLE CS-2000 and CS-3000 comply with the following mandatory and voluntary standards:
- voluntary standards.
Information Technology Equipment Part 1: General Requirements for Safety UL Standard 60950
- Information Technology Equipment, Radio Disturbance (Emissions) Characteristics Limits and Methods of Measurement, IEC/CISPR 22 (EN55022)
- · Information Technology Equipment, Immunity Characteristics Limits and Methods of Measurement, IEC/CISPR 24 (EN55024)
- · Mcacarement, I or Shounications in Medicine) Developed by the American College of Radiology and the National Electrical Manufacturers Association
- 8. Description of Device
REGIUS CONSOLE CS-2000 and CS-3000 are KONICAMINOLTA CR CONSOLE.
REGIUS CONSOLE CS-2000 and CS-3000 control and manage the direct type CR such as INDICD CONCOLL OD 2000 and Sb connected to the Control Unit and the cassette type CR such as REGIUS MODEL 170 and 190 that is connected via the network.
On such as REGION INCES-3000 have the hard disk for storing the digital images.
REGIUS CONSOLE CS-2000 and CS-3000 consist of a console with the touch panel function, a keyboard and a mouse for input, and a barcode scanner.
It can be connected to up multiple direct type CR and multiple cassette type CR, the image read by any reader in the system will be displayed on the REGIUS CONSOLE CS-2000 and
{1}------------------------------------------------
CS-3000 by which the objective cassette was registered. The image read by each reader will OD bood by in real time in synchronization with the reader operation.
REGIUS CONSOLE CS-2000 and CS-3000 process the images received from the reader INDOND OD DONDOID OD 2006 and OSSS of unction, etc., and send them to the connected devices, such as the host computer or the CR printer.
In the case that patient registration is made at each X ray room, the operator can search In the case that patient regionation information via "Examination Search Screen". the patient information of cammiliations a separate reception, the operator can display and On the ocher natia, will and information and Patient Information using "Examination List" screen. In this case, use of an ID Registration System will be necessary.
The console has the following feature.
- le collook has the lonowing reasons.
1. The feature of subscribing patient information (the name, the age, the sexuality, the ID and so on) to display, to choose and to correct or, it is the feature to receive patient and bo only to albums), since toon (the body part, the exposures condition and so on) of the patient and to display, to choose from the hospital-information-system and to correct it.
- 2. The function to enter (the choice) examination information (the body part, the exposures condition and so on).
- Caposares condition to specify the reading condition (the sampling pitch, the reading The fullchon to specify the reading of the image information at the Reader section.
- 4. The function to require the reading of the image data of the Reader section.
- 5. The function to receive image data from the Reader section.
- o. The function to display the image data which was received from the Reader section. Image data is culled to fixed size and is displayed.
- 7. It temporarily saves the image data which was received from the Reader section. The feature
- 8. The function to do an image processing to the image data which was received from the Reader section.
The kind of the image processing
1) Adjusting the Contrast:
Achieve a clearly depicted image (with clear minimum density).
2) F-processing:
// I processins
Highlight fine detail in the image, or enhance detail that has been blurred. These process doses not affect density.
3) E-processing:
// 11 processing due to the wide distribution of the subject.
- 4) H-processing
# II processing is frequency enhancement processing and equalization processing based on multi-resolution analysis.
- 5) Masking
/ Fills in black the area on the frame where the X-ray is not irradiated.
- 6) Rotating/Flipping
Rotate/Reverse the image.
- 7)Re-sampling and Resizing
The function that re-samples and resizes the image data according to need.
8) Stitching
/ Detectings
The function that manually or automatically recognize the long body part and stitch each body part to create a composite image. The image can be divided into several small images before output to a storage device or film printer.
Note) This feature requires cassette type CR in addition connected.
#### 9) Grid Suppression
/ The function suppresses the grid and moire patterns within the image exposed with the grid.
{2}------------------------------------------------
- 9. The function to add digital marker patient information and so on to the image data which was received from the Reader section.
- willch was received from the fictures boom.
10. The function that the printer outputs the image data which was received from the Reader section in the host computer and so on The REGIUS CONSOLE CS-2000 utilizes Discrete Cosine Transform (DCT) as of standard irreversible compression techniques to output the image data to Floppy Disk. Disa.
The REGIUS CONSOLE CS-3000 is not utilized Discrete Cosine Transform (DCT)
Note) Console is designed in the purpose that the radiographer uses and is never used for a diagnosis purpose.
- 11. Portable mode
. The optional the portable mode is the practical use form which PDA(Personal Digital r ne opendial the possible it could carry a bar cord leader department and photography (condition) information from REGIUS CONSOLE CS-2000 and CS-3000 so that a bar cord can register cassette in that place in case of portable photography in the hospitalization ward and so on was used for.
### 9. Diagram of Layout and Interconnections
The figure of the layout and the mutual connection of the system
Image /page/2/Figure/7 description: This image shows a diagram of the CS-2/3(REGIUS CONSOLE CS-2000/3000) system. The system is connected to HIS, RIS, Printer, PACS, Viewer, Image Reader (REGIUS MODEL 170), and Konicaminolta PDAs via Ethernet (IEEE802.3*). The system is also connected to Image Reader (REGIUS MODEL 370) via Image Reader I/F (IEEE1394).
*6
{3}------------------------------------------------
- * 1 DICOM Modality Worklist Information Model FIND
- *2 DICOM Basic Grayscale Print Management Meta SOP Class
- *3 DICOM Computed Radiography Image Storage DICOM Storage Commitment Push Model SOP Class
- *4 Konicaminolta original protocol
- *5 Konicaminolta original protocol
- *6 Konicaminolta original protocol
- 10. Safety Information
REGIUS CONSOLE CS 2000 and CS 3000 introduce no new safety and efficacy issues other than those already identified with the predicate device. The results of a hazard analysis, than those an each fractive measure taken indicate that the device is of minor contention will as apper the May 29, 1998 issue of the "Guidance for the Content of Premark Submissions for Software Contained in Medical Devices".
- 11. Substantial Equivalence to Predicate Device
The REGIUS CONSOLE CS-2000 and CS-3000 are substantially equivalent to Fuji CR Console Plus (Flash Plus IIP), 510(k) number: K041990.
Comparison of the principal characteristics of the two devices which are pertinent to Specification performance is shown below.
- 1) Hardware
| Feature | REGIUS CONSOLE<br>CS-3000 | REGIUS CONSOLE<br>CS-2000 | Fuji Flash Plus IIP | |
|-----------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|--------------------------------------|
| Minimum Basic<br>Computer<br>Configuration | Computer "Off the Shelf"<br>· Desktop or Tower<br>· CPU: Pentium 4<br>· Bus: PCI<br>· RAM: 1GB<br>· Hard Drive: 80GB<br>· Floppy Drive: 3.5"<br>· CD-ROM<br>· Keyboard<br>· Mouse<br>· Barcode scanner | Computer "Off the Shelf"<br>· Desktop or Tower<br>· CPU: Pentium 4<br>· Bus: PCI<br>· RAM: 512MB<br>· Hard Drive: 80GB<br>· Floppy Drive: 3.5"<br>· CD-ROM<br>· Keyboard<br>· Mouse<br>· Barcode scanner | Computer "Off the Shelf"<br>· Desktop or Tower<br>· CPU: Pentium 4<br>· Bus: ISA/PCI<br>· RAM: 512MB<br>· Hard Drive: 40GB<br>· Floppy Drive: 3.5"<br>· CD-ROM<br>· Keyboard<br>· Mouse<br>· Barcode scanner | |
| Operating System<br>Software | Microsoft Windows XP | Microsoft Windows XP | Microsoft Windows 2000<br>or Windows XP | |
| Ethernet Capability &<br>Type | Yes:LAN | Yes:LAN | Yes:LAN | |
| Image transfer | via DICOM 3.0 &<br>Via proprietary protocol | via DICOM 3.0 &<br>Via proprietary protocol | via DICOM 3.0 &<br>Via Fuji DMS Network | |
| Image Display | 15" color 1MP LCD<br>with Touch screen | 15" color 1MP LCD<br>with Touch screen | 19" color 2MP LCD<br>with Touch screen | |
| Image processing<br>functions | Yes, enhanced | Yes, enhanced | Yes, enhanced | |
| Image viewing &<br>orientation functions | Yes, enhanced | Yes, enhanced | Yes, enhanced | |
| Connects to<br>Konicaminolta<br>Image Readers | Yes | Yes | No | |
| Connects to<br>Konicaminolta<br>Image<br>Recorders(Printers) | Yes | Yes | Yes | |
| Connects to<br>Konicaminolta<br>PDAs(The option for<br>portable mode) | Yes | No | | |
| Feature | REGIUS<br>CONSOLE<br>CS-3000 | REGIUS<br>CONSOLE<br>CS-2000 | Fuji Flash Plus IIP | |
| Image<br>processing | a. F-processing<br>F-Processing is a form of image processing which<br>modifies image spatial frequency characteristics, so<br>that structures of body parts are displayed more<br>sharply. | | a. Gradation/Edge enhancement<br>Along with gradation, edge enhancement as<br>well as DRC Fuji image processing makes image<br>quality consistently good, so the technologist<br>spends less time manipulating the image. | Same as<br>the<br>approved<br>device |
| | b. E-processing<br>E-processing allows an image with a wide dynamic<br>range to be converted to one with a smaller<br>dynamic range which is easier to view. | | b. DRC (Dynamic Range Control)<br>DRC improves visualization of areas with<br>different densities in the same image. | Same as<br>the<br>approved<br>device |
| | c. H-processing<br>H-Processing is the method of frequency<br>Processing that uses the resolution of the image<br>in multi resolution space. This adjusts the<br>sharpness of the image and compress the dynamic<br>range. | | c. MFP(Multi-objective Frequency Processing)<br>MFP enables enhancement of both small and<br>large structures at the same time as well as<br>better visualization of areas with different<br>densities. | Same as<br>the<br>approved<br>device |
| | d. Grid Suppression<br>The function suppresses the grid and moire<br>patterns within the image exposed with the grid,<br>processing by filtering. | | d. GPR (Grid Pattern Removal)<br>GPR removes stationary grid pattern to<br>suppress moire patterns within an image. | Same as<br>the<br>approved<br>device |
| | e. Masking<br>Masking blacks out areas outside the field of X-ray<br>exposure on the image. | | | Additional<br>feature |
| | f. Re-sampling and Resizing<br>The functions change the resolution of the image,<br>using digital image interpolation according to need. | | | Additional<br>feature |
| | g. Stitching<br>Stitching assembles the composite image from the<br>images read out the photostimulated luminescence<br>plates which had been positioned such that the two<br>adjacent plates overlap each other at the exposure.<br>This manually or automatically adjusts the image<br>positions which are matched with each other. | | | Additional<br>feature |
{4}------------------------------------------------
## 2) Software
:
{5}------------------------------------------------
Image /page/5/Picture/1 description: The image shows the logo for the U.S. Department of Health and Human Services. The logo is circular and contains the text "DEPARTMENT OF HEALTH & HUMAN SERVICES - USA" around the perimeter. Inside the circle is an abstract symbol that resembles a bird or a stylized human figure.
Food and Drug Administration 9200 Corporate Boulevard Rockville MD 20850
Konica Minolta Medical & Graphic, Inc. % Mr. Shinishi Yamanaka Safety Department Cosmos Corporation 319 Akeno, Obata-cho, Watarai-gun Mie-ken, 519-05 JAPAN
Re: K051523
JUL 2 0 2005
Trade/Device Name: Medical Image Processing Workstation, REGIUS CS-2000/CS-3000 Regulation Number: 21 CFR 892.2050 Regulation Name: Picture archiving and communications system Regulatory Class: II Product Code: LLZ Dated: June 6, 2005 Received: June 14, 2005
Dear Mr. Yamanaka:
We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food, Drug, and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, not the device, subject to the general controls provisions of the Act. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration.
If your device is classified (see above) into either class II (Special Controls) or class III (Premarket Approval), it may be subject to such additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with the Act's requirements, including, but not limited to registration and listing (21 CFR Port 807) vian ing (21 CFR Part 801); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.
{6}------------------------------------------------
This letter will allow you to begin marketing your device as described in your Section 510(k) premarket notification. The FDA finding of substantial equivalence of your device to a legally marketed predicate device results in a classification for your device and thus, permits your device to proceed to the market.
If you desire specific advice for your device on our labeling regulation (21 CFR Part 801), please contact the Office of Compliance at one of the following numbers, based on the regulation number at the top of this letter:
| 21 CFR 876.xxxx | (Gastroenterology/Renal/Urology) | 240-276-0115 |
|-----------------|----------------------------------|--------------|
| 21 CFR 884.xxxx | (Obstetrics/Gynecology) | 240-276-0115 |
| 21 CFR 892.xxxx | (Radiology) | 240-276-0120 |
| Other | | 240-276-0100 |
Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR 807.97). You may obtain other general information on your responsibilities under the Act from the Division of Small Manufacturers, International and Consumer Assistance at its toll-free number (800) 638-2041 or (301) 443-6597 or at its Internet address http://www.fda.gov/cdrh/industry/support/index.html.
Sincerely yours,
Nancy C. Brogdon
Nancy C. Brogdon Director, Division of Reproductive, Abdominal, and Radiological Devices Office of Device Evaluation Center for Devices and Radiological Health
Enclosure
{7}------------------------------------------------
# Indications for Use
510(k) Number (if known) : Қ ( 6 ﻛ
Device Name
- Medical Image Processing Workstation, REGIUS CS-2000/ .. CS-3000
Indications For Use:
Receive and process electronic images of patients. The REGIUS CS-2000/CS-3000 is NOT intended for use with digital mammography system.
| Prescription Use | |
|-----------------------------|--|
| (Part 21 CFR 801 Subpart D) | |
AND/OR
Over-The-Counter Use (21 CFR 807 Subpart C)
(PLEASE DO NOT WRITE BELOW THIS LINE-CONTINUE ON ANOTHER PAGE IF NEEDED)
Concurrence of CDRH, Office of Device Evaluation (ODE)
Nancy Brogdon
(Division Sign-Off)
of Reproductive, Abdominal, diological Devices 510(k) Number
Page 1 of
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.