K100092 · Best Nomos · MUJ · Feb 23, 2010 · Radiology
Device Facts
Record ID
K100092
Device Name
CORVUS RADIATION THERAPY PLANNING SYSTEM
Applicant
Best Nomos
Product Code
MUJ · Radiology
Decision Date
Feb 23, 2010
Decision
SESE
Submission Type
Special
Regulation
21 CFR 892.5050
Device Class
Class 2
Attributes
Software as a Medical Device
Indications for Use
CORVUS is intended for use as a planning tool for conformal radiation therapy. Using operator-supplied input and patient scans, it creates a plan for treatment delivery systems and generates a set of beam weights that, when applied to a compatible system, facilitates delivery of an intensity-modulated 3D conformal radiation therapy treatment. CORVUS is intended only to suggest a delivery plan. It is the physician's responsibility to verify that the dose distributions which would result from plan implementation are appropriate for a particular patient. The CORVUS system is intended to be used as an integrated system with a modulating device for planning and delivery of conformal radiation therapy. The modulating device can be the NOMOS MIMIC, nomosSTAT MLC, or a supported MLC. CORVUS produces radiation fields which are modulated to conform to the projected tumor volume plus margins. The system tries to achieve target goals while sparing sensitive structures.
Device Story
CORVUS is a semi-automatic radiation therapy treatment planning system. It ingests patient scans and operator-supplied inputs to generate 3D conformal radiation therapy plans. The system uses optimization methods—simulated annealing and gradient descent—to calculate beam weights for pencil beams, ensuring target dose coverage while sparing sensitive structures. It supports volumetric dose calculation using finite size pencil beam (FSPB) and Lateral Disequilibrium Inclusive (LDI) algorithms. Operated by medical physicists, dosimetrists, and radiation oncologists in clinical settings, the system provides a guide for delivery via compatible MLCs (e.g., NOMOS MIMIC, nomosSTAT, Siemens ONCOR). The 'ActiveRx' feature allows clinicians to interactively adjust isodose lines and constraints before final dose calculation. The output is a treatment plan that the physician must review and approve. By optimizing beam weights and gantry positions, the device facilitates precise dose delivery, potentially improving tumor control while minimizing damage to healthy tissue.
Clinical Evidence
Bench testing only. No clinical trials. Validation included 53 comparisons with measurements or Monte-Carlo calculations and 62 system tests. Accuracy of the LDI algorithm was verified against film, MOSFETs, and ion-chambers using phantoms simulating low-density regions. Plan quality was evaluated by qualified personnel familiar with inverse treatment planning systems. All anomalies were resolved, and performance was confirmed to be conformant with requirements.
Technological Characteristics
Radiation therapy planning system; Intel-based Mac Pro hardware; supports DICOM-compatible imaging inputs. Algorithms: Finite Size Pencil Beam (FSPB) and Lateral Disequilibrium Inclusive (LDI) heterogeneity correction. Connectivity: Standalone workstation (no longer requires remote compute engine). Interface: WACOM CINTIQ 21UX tablet support. Software: MacOS X 10.5.4 based. Optimization: Simulated annealing and gradient descent.
Indications for Use
Indicated for medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray) external beam radiation therapy. Valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are not supported.
Regulatory Classification
Identification
A medical charged-particle radiation therapy system is a device that produces by acceleration high energy charged particles (e.g., electrons and protons) intended for use in radiation therapy. This generic type of device may include signal analysis and display equipment, patient and equipment supports, treatment planning computer programs, component parts, and accessories.
{0}------------------------------------------------
Kl.00092
Section 2 Special 510(k) Summary Prepared January 7, 2010
FEB 2 3 2010
## 2.1 SUBMITTERS NAME
Best NOMOS
#### 2.2 ADDRESS
One Best Drive Pittsburgh, PA 15202
#### 2.3 CONTACT
Chi Palko Phone: (412) 312-6744 Fax: (412) 312-6701
#### 2,4 DEVICE NAME
Device Trade Name: CORVUS Common/Classification Name: Radiation Therapy Treatment Planning System
## 2.5 ESTABLISHMENT REGISTRATION NUMBER
Corporate Office Registration Number: 2434141 Manufacturing and Packaging Registration Number: 2434141
#### 2.6 DEVICE CLASS
Class II Panel: Radiology Product Code: 90-MUJ Regulation Number: 21 CFR 892.5050
## 2.7 LEVEL OF CONCERN
#### Major Level of Concern
We believe the level of concern is Major for CORVUS 09. CORVUS 09 is an accessory to a medical device that has a Major Level of Concern such as delivery systems which use modulated radiation therapy for delivery and planning.
{1}------------------------------------------------
The copy of the complete decision making process for this conclusion is included in Section 13.1.
#### 2.8 STATEMENT OF INDICATIONS FOR USE
The intended use and indications for use for the CORVUS Radiation Therapy Planning System have NOT changed as a result of this modification.
#### Intended Use
CORVUS is intended for use as a planning tool for conformal radiation therapy. Using operator-supplied input and patient scans, it creates a plan for treatment delivery systems and generates a set of beam weights that, when applied to a compatible system, facilitates delivery of an intensity-modulated 3D conformal radiation therapy treatment. CORVUS is intended only to suggest a delivery plan. It is the physician's responsibility to verify that the dose distributions which would result from plan implementation are appropriate for a particular patient.
The CORVUS system is intended to be used as an integrated system with a modulating device for planning and delivery of conformal radiation therapy. The modulating device can be the NOMOS MIMIC, nomosSTAT MLC, or a supported MLC. CORVUS produces radiation fields which are modulated to conform to the projected tumor volume plus margins. The system tries to achieve target goals while sparing sensitive structures.
#### Indications for Use
The CORVUS system is a radiation treatment planning package designed to allow medical physicists, dosimetrists, and radiation oncologists to create conformal treatment plans using photon (x-ray) external beam radiation therapy. The treatment plans generated by CORVUS are based upon treatment machine-specific data and are intended to provide a guide to delivering external beam radiation therapy which conforms to the target volume defined by the radiation oncologist.
The CORVUS system is valid for use only with external beam photon therapy; calculations for electrons and intracavity sources (Brachytherapy) are NOT supported,
#### 2.9 DEVICE DESCRIPTION
CORVUS is a semi-automatic planning system: rather than simply verifying a userdesigned plan, the system itself suggests a plan. A clinician then reviews and approves the plan.
CORVUS is designed to generate plans for treatment delivery systems that can create multiple radiation patterns composed of pencil beams on which the intensity can be individually controlled. The treatment beams are weighted so that when they are projected into the treatment space they superimpose to give the desired dose distribution.
{2}------------------------------------------------
Each radiation field is generated using one of several optimization methods provided with the system, including simulated annealing and gradient descent.
The treatment beams are set not only to deliver the prescribed dose to the identified target volume, but also to keep the dose to other sensitive volumes below user-defined limits. Planning is done volumetrically: the beam weights for treating the entire target volume are generated simultaneously. The dose matrix is volumetric. The dose to each point is calculated to be that received from all beams and from all gantry angles. Dosage is calculated using a finite size pencil beam (FSPB) algorithm based on the beam characterization of clinically measured data. The degree to which a treatment plan is optimized is determined in part by constraints placed on the planning algorithm. The user has direct control over these constraints, which include dose goals to the target structures, dose limits to the sensitive structures, and the specification of arcs or fixed gantry positions in the treatment plan.
CORVUS treatment plans need not have the isocenter located within the target volume. An unlimited number of targets falling within the treatment volume can be planned for at the same time. Dose may be prescribed for up to 32 structures, 29 of them userselectable, any number of which may be separate targets or radiation-sensitive structures. Each structure can have a separate dose prescription.
## 2.10 PREDICATE DEVICE INFORMATION
The CORVUS 09 system is substantially equivalent to the CORVUS 5.0M (K032209). The CORVUS 5.0M was determined to be substantially equivalent to its predicate device as of August 2003. 1
The fundamental scientific technology for the CORVUS 5.0M and CORVUS 09 systems has not changed. The intended use of the device has not changed. Based upon the performance testing results for CORVUS 09, the system raises no new issues of safety or effectiveness.
#### 2.11 COMPARISON TO THE PREDICATE DEVICE
This section describes the incremental changes from CORVUS 5.0M to CORVUS 09.
*Note that for marketing purposes the CORVUS 6.4 version was renamed as CORVUS 08 and CORVUS 7.0 was renamed to CORVUS 09, however internally either may be referred to as its version level.*
#### Hardware and Operating System
CORVUS 5.0M
- Operating System: MacOS X 10.2.6 operating system. .
- Mac-based Power Pro G4 hardware (based on Apple Power ● Workstation: Mac G4 Dual 1.42 Ghz CPU)
{3}------------------------------------------------
- . Server: OmnitTech Percheron Intel 860 Dual 2.4 Ghz
- . Printer: Xerox Tektronix Phaser 6200 Laser Printer
CORVUS 09
- Operating System: . MacOS X 10.5.4 operating system
- Workstation: Intel-based Mac Pro hardware system (based on two 3.2GHz Quad-Core Intel Xeon (8-core)CPUs).
- Physician's Review Workstation: Intel-based iMac hardware system (based on . 2.03 Ghz Intel core 2 Duo CPU)
- WACOM CINTIQ 21UX Interactive Pen Display, ● WACOM Tablet: Graphics Tablet
Note that the development of more powerful computers has climinated the need for a separate server computer solely for planning, like used in CORVUS 5.0M. The more powerful CORVUS 09 workstation completes the planning locally (on the Workstation) and does not require a Remote Compute Engine (RCE). The Physician's Review Workstation released in CORVUS 09 allows remote plan review through a limited CORVUS application. The printer is no longer included as a component of the device.
#### CORVUS Software Features
- CORVUS 09 contains the Active Rx feature addition which was introduced in . CORVUS version 6.0.
- ActiveRx is an IMRT optimizer where the user plays an integral.role in o the optimization process with controls to influence and evaluate the dosimetric tradeoffs before plan delivery. ActiveRx allows CORVUS users to make certain kinds of adjustments to treatment plans in DisplayResults mode by directly manipulating some isodose lines, CDVH curves, delivery efficiency characteristics and dose limit constraints while maintaining physically deliverable plans. ActiveRx displays updated isodose lines, CDVHs, statistics, and treatment plans as the user makes changes. Prior to approval, the system requires the user to submit a plan for final dose calculation.
- CORVUS 09 contains support for formulation and plan output for treatment o delivery with the Siemens ONCOR 82 Leaf MLC. This was introduced in CORVUS version 6.0.
- CORVUS 09 contains the feature to use a 2D array (such as MapCheck or . MatriXX) to create a Hybrid Plan which is intended to be delivered during the OA process. This involves reorienting beams to gantry vertical in order adapt to the orientation dependence of the 2D array. This was introduced in CORVUS 6.2.
- . CORVUS 09 contains support for nomosSTAT (K060859) which was introduced in CORVUS version 6.3.
{4}------------------------------------------------
- CORVUS 09 supports the optional use of WACOM CINTIQ 21UX electronic . drawing tablet as an additional interface. This was introduced in CORVUS OF
- CORVUS 09 introduces improved dose calculation (Lateral Disequilibrium . Inclusive (LDI) heterogeneity correction algorithm) applicable for all iterations of the optimization thereby improving treatment plans in low-density regions such as the lungs. This option was not available in any prior software version.
- CORVUS 09 introduces improved leakage calculation for Varian and Siemens . Multi-Leaf Collimators (MLCs). Prior versions of CORVUS incorporated this type of leakage calculation for Elekta MLCs.
While considering these differences it is important to note that CORVUS 09 prepares equivalent treatment plans outside of low density regions as produced by CORVUS 5.0M. The process through which CORVUS accepts image and user input, ortimizes plans through applying simulated annealing or gradient descent to cost functions, forms and displays results, and performs all other reatment planning steps have not changed. The algorithms and methods used by CORVUS 5.0M are equivalent to CORVUS 09 except for the inclusion of the improved Lateral Disequilibrium Inclusive (DI) heterogeneity correction algorithm. This has improved dose calculation in low-density regions. CORVUS 09 also improves the leakage calculation in Multi-Leaf Collimators (MLCs). The process flow and user interface remain the same for CORVUS except for the inclusion of a user optional post optimization interactive tool (Active Rx).
The incremental changes since the latest CORVUS 510(k) were reviewed. We believe that this Special 510k submission is warranted for inclusion of the new Lateral Disequilibrium Inclusive (LDI) dose calculation in CORRVUS 09.
Other incremental changes are reviewed. NO particular change NOR the sum of the incremental changes between CORVUS 5.0M and CORVUS 08 could significantly affect the safety or effectiveness of the device.
- Active Rx did not change the performance specifications in a manner that could . affect the safety and effectiveness of the device. The final dose calculation upon which clinical users approve the plan was not substantially modified. Additionally, the quality assurance techniques that are used clinically such as phantom delivery still apply.
- The Siemens ONCOR 82 Leaf MLC support did not affect the safety and . effectiveness of the device. This support is similar to prior supported MLCs with analogous design control. Additionally, the quality assurance techniques that are used clinically such as phantom delivery still apply.
- Support of a 2D array (such as MapCheck® or MatriXX™) to create a Hybrid . Plan is substantially similar to existing QA plan functionality.
- The support of nomosSTAT (K060859) is substantially equivalent to the support . of the MIMiC(K940412) and Autocrane(K013036) systems by CORVUS 5.0M.
{5}------------------------------------------------
- The support of the WACOM CINTIQ 21UX electronic drawing tablet did not . affect the safety or effectiveness of the device.
### Performance Testing Data
A total of 53 comparisons with measurement or comparisons with Monte-Carlo were completed. Equivalence Tests to Confirm Equivalence of the CORVUS-software used for validation as compared with the final version are provided in the verification were completed. Anomalies identified in the measured deta were investigated and satisfactory resolutions were obtained. The accuracy of the system was confirmed and found to be conformant with our requirements. The plan quality was evaluated and found substantially equivalent. (Refer to Appendix F CORVUS 09 Total Validation Report for further details)
A total of 62 system tests passed the criteria. Module tests were completed for new or modified code. Defect resolutions were verified by an independent evaluator. New or modified code was evaluated by an independent reviewer. Final high level run-through tests were performed to confirm the final functionality of CORVUS. (Refer to Appeadix F for the CORVUS 09 Total Verification Summary Report for further details)
CORVUS 09 supports two heterogeneous dose calculation algorithms and introduces improvements to leakage calculation for the Varian and Siemens MLC's:
- The Effective Path Length (EPL) algorithm which has existed in all versions of . CORVUS including the predicate version 5.0M and up to version 09. Dosimetric validation of CORVUS' EPL algorithm was performed by comparing it with film measurements, MOSFETS and ion-chambers (See CORVUS 09 Total Validation Report for more details). Treatment plan quality validation was accomplished by comparing.treatment plans generated by CORVUS 09 using the EPL algorithm with those generated by a prior version (CORVUS 08) and determined to be substantially equivalent. This was expected as the EPL algorithm has not undergone any substantial changes since CORVUS 5.0M.
- The Lateral Disequilibrium Inclusive (LDI) algorithm which is introduced in t CORVUS 09. Dosimetric validation of CORVUS' LDI algorithm was performed similarly by comparing it with film measurements, Mostets and ion-chambers. Treatment plan quality validation using the new LDI algorithm was domeby qualified personnel who are familiar with using an inverse treatment planning system in a clinical setting. (See Appendix F for the CORVUS 09 Total Validation Report for more details)
- Performance testing for the improvements to leakage calculation for the Varian . and Siemens MLCs were conducted and found to be conformant with our requirements. (See Appendix F for the CORVUS 09 Total Validation Report for more details)
{6}------------------------------------------------
## 2.12 SUMMARY OF DESIGN CONTROL ACTIVITIES
The new lateral disequilibrium inclusive pencil-beam algorithm provides improved performance in low density regions such as those found in the lung, while it does not constitute a change of fundamental scientific technology since the planning system. creates treatment plans which are confined to conformal radiation therapy using platon radiation as in prior software versions.
The modifications in CORVUS 09 are appropriate for reliance on NOMOS' design control process. The accuracy of the change is evaluated through comparison with medical physics measurements including film measurement, ion chamber measurement, MOSFET measurement, as well as Monte-Carlo calculation using a phantom designed to simulate low-density regions of the patient which are well known to medical plugicious skilled in the art of conformal radiation therapy. The quality of the treatment plyon is evaluated based upon clinical criteria which are associated with conformal the pany similarly to evaluations which occurred in prior product versions which are well known to medical physicists. Refer to Section 6 for the Declaration of Conformity.
{7}------------------------------------------------
Image /page/7/Picture/1 description: The image shows the logo for the U.S. Department of Health and Human Services. The logo consists of a stylized eagle with three tail feathers, representing the department's commitment to health, human services, and well-being. The eagle is encircled by the text "DEPARTMENT OF HEALTH & HUMAN SERVICES USA" in a circular arrangement.
Food and Drug Administration 10903 New Hampshire Avenue Document Control Room -WO66-G609 Silver Spring, MD 20993-0002
# MAR 1 1 2010
Mr. Chi Palko Quality Manager Best NOMOS 1 Best Drive PITTSBURGH, PA 15202
Re: K100092
Trade/Device Name: CORVUS Radiation Therapy Treatment Planning System (Model: CORVUS 09) Regulation Number: 21 CFR 892.5050 Regulation Name: Radiation Therapy Treatment Planning System Regulatory Class: II Product Code: MUJ and IYE Dated: February 23, 2010 Received: February 23, 2010
Dear Mr. Palko:
This letter corrects our substantially equivalent letter of February 23, 2010. 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 (PMA). You may, therefore, market the device, subject to the general controls provisions of 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 (PMA), it may be subject to 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
{8}------------------------------------------------
or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR Part 807); labeling (21 CFR Part 801); medical device reporting (reporting of medical device-related adverse events) (21 CFR 803); 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.
If you desire specific advice for your device on our labeling regulation (21 CFR Part 801), please go to http://www.fda.gov/AboutFDA/CentersOffices/CDRH/CDRHOffices/ucm115809.htm for the Center for Devices and Radiological Health's (CDRH's) Office of Compliance. Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to
http://www.fda.gov/MedicalDevices/Safety/ReportalProblem/default.htm for the CDRH's Office of Surveillance and Biometrics/Division of Postmarket Surveillance.
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) 796-7100 or at its Internet address http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/default.htm.
Sincerely yours,
NuefflR
Donald J. St.Pierre Acting Director Division of Radiological Devices Office of In Vitro Diagnostic Device Evaluation and Safety Center for Devices and Radiological Health
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.