Clinical data from ventilated patients in ICU/hospital units; Peer-reviewed clinical literature (Banner et al., 2008)
Retrospective and clinical study data were used to validate the non-invasive WOB/min algorithm against invasive esophageal pressure measurements and to test the clinical agreement of the PS/V Advisor rule-based system.
Work of breathing; Clinical validation; ICU patients; Algorithm validation
Clinical Evidence
Study Design
Population
Comparator
Key Endpoints
WOB/min algorithm validation; Comparison of non-invasive measures to invasive esophageal balloon measurements
Ventilated adults in the ICU (MICU, CICU, SICU, burn unit, step-down unit) breathing spontaneously on pressure support
PS/V Advisor clinical studies (including Banner et al., 2008); Clinical testing of rule-based advisory system (one study blinded, one unblinded)
Mechanically ventilated adult patients
Not applicable for this study
Clinician agreement with PS Advisor recommendations
AI Performance
Output
Algorithm
Acceptance
Observed
Dev DS
Dev Readers
Test DS
Test Readers
Work of breathing per minute
Artificial Neural Net
—
bias and precision of 0.84 ±2.2 J/min
—
—
Clinical study of ventilated adults in the ICU (MICU, CICU, SICU, burn unit, step-down unit) breathing spontaneously on pressure support ventilation.
—
Ventilator support advice
Fuzzy-logic rule based system
—
—
—
—
Two separate clinical studies involving clinician agreement with PS Advisor recommendations.
>1 (clinicians)
Indications for Use
The intended use of the Philips NM3 Respiratory Profile Monitor, Model 7900, is to provide: - cardiac output monitoring via the method of partial rebreathing in adult patients receiving mechanical ventilation during general anesthesia and in the intensive care unit (ICU). - spirometric, and carbon dioxide monitoring in neonatal, pediatric and adult patients during general anesthesia and in the intensive care unit (ICU) and the emergency department (ED). Separate combination CO2/flow sensors are provided for adult, pediatric and neonatal use. - continuous, non-invasive monitoring of functional arterial oxygen saturation and pulse rate in neonatal, pediatric and adult patients during both no motion and motion conditions and for patients who are well or poorly perfused during general anesthesia and in the intensive care unit (ICU) and the emergency department (ED). The intended use of the VentAssist software option is to provide: - non-invasive monitoring of work of breathing per minute in adult patients receiving pressure support mechanical ventilation. - on-demand advice in mechanically ventilated adult patients as prescribed by the caregiver regarding (a) modifications to the current pressure support settings in order to assess the work of breathing and breathing pattern and (b) modifications to ventilation in order to maintain end-tidal CO2 in a range determined by the physician. The patients should be hemodynamically stable and must be breathing spontaneously.
Device Story
Philips NM3 Respiratory Profile Monitor (Model 7900) is a multiparameter monitor for ICU, anesthesia, and ER use. It processes inputs from combination CO2/flow sensors, pulse oximetry sensors, and sidestream gas modules. The device provides real-time monitoring of respiratory flow, airway pressure, CO2, SpO2, pulse rate, and cardiac output (via partial rebreathing). The VentAssist software option adds decision support for mechanical ventilation; it calculates work of breathing per minute (WOB/min) using an Artificial Neural Net (ANN) and provides on-demand advice for pressure support settings via a fuzzy-logic rule-based system. Clinicians view parameters and advice on a dedicated screen; they may accept or reject recommendations to adjust ventilator settings. This system aims to optimize ventilatory support, reduce excessive work of breathing, and maintain target end-tidal CO2 levels in spontaneously breathing, hemodynamically stable adult patients.
Clinical Evidence
Bench testing validated the WOB/min front-end against the Bicore CP-100 and confirmed algorithm implementation. Clinical validation of the PS/V Advisor rule set involved two studies: one assessing clinician agreement with recommendations (published in Banner et al., Chest, 2008) and a second blinded study. WOB/min accuracy demonstrated bias and precision of 0.84 ± 2.2 J/min compared to invasive esophageal balloon measurements in spontaneously breathing ICU patients.
Technological Characteristics
Multiparameter monitor utilizing mainstream/sidestream gas analysis (infrared), pulse oximetry (Masimo Rainbow Set), and differential pressure flow sensing. VentAssist software implements WOB/min calculation via Artificial Neural Net (ANN) and advisory logic via fuzzy-logic rule-based system. Connectivity includes sensor interfaces for adult/pediatric/neonatal populations. Embedded system architecture.
Indications for Use
Indicated for neonatal, pediatric, and adult patients in ICU, anesthesia, or ED settings requiring respiratory monitoring (CO2, flow, pulse oximetry). VentAssist software indicated for hemodynamically stable, spontaneously breathing adult patients on pressure support mechanical ventilation.
Regulatory Classification
Identification
A monitoring spirometer is a device used to measure continuously a patient's tidal volume (volume of gas inhaled by the patient during each respiration cycle) or minute volume (the tidal volume multiplied by the rate of respiration for 1 minute) for the evaluation of the patient's ventilatory status.
{0}------------------------------------------------
KI03578
## 510(k) Summary
( Refer to 21 C.F.R. § 807.92 )
#### Submitted by a.
JUN - 2 2011
Respironics Novametrix LLC. 5 Technology Drive Wallingford, CT 06492
#### b. Contact Person
Kevin R. Mader Senior Quality and Requlatory Manager Phone: (203) 697-6466 Fax: (203) 284-0753
- C. Device Name
Proprietary/Trade Names:
Common Name:
Philips NM3 Respiratory Profile Monitor with VentAssist, Model 7900
multiparameter monitor (monitoring spirometer, CO2 monitor, pulse oximeter and cardiac output monitor with partial rebreathing valve).
Classification:
Class II, 21 C.F.R. 868.1850, 868.1400, 870.2700, 868.5675
#### Predicate Devices ರ.
Substantial equivalence to the following legally marketed predicate devices with the same or similar indications for use has been demonstrated by a comparison of product features as described in the labeling and promotional literature for predicate devices. The predicate devices are as follows:
| 1. Philips NM3 Respiratory Profile Monitor | K091459 |
|---------------------------------------------------|---------|
| 2. Bicore CP-100 Cardiopulmonary Monitor | K900696 |
| 3. Puritan Bennett Model 840 Ventilator with PAV+ | K053388 |
| 4. Dräger EvitaXL with Option SmartCare | K051263 |
#### Device Description e.
The NM3 monitor with VentAssist is intended for non-invasive monitoring of the inspired and expired airflow and airway pressure of intensive care unit (ICU), anesthesia and emergency room (ER) patients, as well as capnography and pulse oximetry in all of these clinical settings. It is intended to serve all of the same purposes as the flow, carbon dioxide, pulse oximetry, and cardiac output monitoring components of the predicate NM3 monitor with the addition of the optional VentAssist software.
Combination CO2 adapter/flow sensors (neonatal, pediatric, adult), combination adult CO2 adapter/flow sensors with a partial rebreathing valve and flow sensors (infant/neonatal, pediatric/adult) are connected with a male pneumatic connector to the NM3 monitor. Sidestream airway adapters and nasal cannulas are available which are connected with a sample cell connector to a receptacle on the LoFlo Module which can be interfaced to the
{1}------------------------------------------------
NM3 monitor. All of these sensors are already legally marketed as accessories of 510(k) cleared Respironics-Novametrix NM3 monitor. The pulse oximetry sensors are connected to the NM3 monitor via a connector on the front panel of the monitor. All of the pulse oximetry sensors are already legally marketed as accessories of the 510(k) cleared NM3 monitor and Masimo predicate devices.
The principal function of the flow portion of combination sensors and flow sensors is to provide a differential pressure signal related to flow and airway pressure relative to atmospheric pressure. These sensors are often placed in the breathing circuit between the endotracheal tube and the ventilator circuit Y piece and may also be used in conjunction with a face mask or mouthpiece. The flow measurement portion of the NM3 monitor consists of a microprocessor-based data acquisition system that measures flow, and pressure and interfaces with a Capnostat 5 CO2 sensor. The CO2 airway adapter portion of the combination sensors, allow the Respironics-Novametrix CO2 mainstream gas sensor, the Capnostal® 5, to attach to it and measure the concentration of CO2 in the airway using infrared technology. When CO2 measurements are combined with airway flow and volume measurements, other parameters such as CO2 production and dead space can be calculated in all patient populations. The Capnostat 5 sensor as a mainstream gas analyzer includes a sample cell positioned in the breathing circuit through which a patient's inspiratory and expiratory gases flow. The LoFlo module, a sidestream type of gas analyzer, samples gases at 50 ml/min from a sampling port in an adapter placed in a breathing circuit or from a nasal or oral cannula. The gas then passes through a sampling tube to the sample cell, where the gas components are measured. The combination adult CO2 adapter/flow sensors with a partial rebreathing valve with periodic activation of the rebreathing valve allow pulmonary capillary blood flow and cardiac output to be calculated using the differential Fick method.
The VentAssist software option comprises a new screen with a soft key that provides ondemand ventilator-independent open-loop advice with respect to the level of pressure support and ventilation. As an advisory system, the clinician can choose to accept or reject the advice, alleviating any issues of safety and effectiveness. Additionally, an improved method for the calculation of plateau pressure has been included, as well as a new calculated parameter for work of breathing (WOB). The WOB parameter facilitates the goal of reducing excessive work of breathing per minute, or power of breathing (WOB/min), for mechanical ventilatory support in patients with respiratory failure. The WOB/min parameter is implemented using an Artificial Neural Net (ANN) and is used by the PSV Advisor software. The VentAssist PS/V Advisor is a rule based system which provides on-demand advice for the setting of the PSV level and ventilator support levels, based upon WOB/min, breathing frequency, tidal volume, ideal body weight, and end-tidal CO2. The advice is based upon a set of logic rules developed and refined in conjunction with experienced critical care clinicians at teaching University hospitals. Decision support advice offered by the VentAssist software is available during monitoring of adult patients. The monitor uses the sensor-type (adult, pediatric, or neonate), as well as patient data entered into the monitor, to enable appropriate features.
#### f. Intended Use
The intended use of the Philips NM3 Respiratory Profile Monitor, Model 7900, is to provide:
- cardiac output monitoring via the method of partial rebreathing in adult patients . receiving mechanical ventilation during general anesthesia and in the intensive care unit (ICU).
- spirometric, and carbon dioxide monitoring in neonatal, pediatric and adult patients . during general anesthesia and in the intensive care unit (ICU) and the emergency
{2}------------------------------------------------
department (ED). Separate combination CO2/flow sensors are provided for adult, pediatric and neonatal use.
- continuous, non-invasive monitoring of functional arterial oxvgen saturation and pulse . rate in neonatal, pediatric and adult patients during both no motion and motion conditions and for patients who are well or poorly perfused during general anesthesia and in the intensive care unit (ICU) and the emergency department (ED).
The intended use of the VentAssist software option is to provide:
- non-invasive monitoring of work of breathing per minute in adult patients receiving . pressure support mechanical ventilation.
- on-demand advice in mechanically ventilated adult patients as prescribed by the . caregiver regarding (a) modifications to the current pressure support settings in order to assess the work of breathing and breathing pattern and (b) modifications to ventilation in order to maintain end-tidal CO2 in a range determined by the physician. The patients should be hemodynamically stable and must be breathing spontaneously.
## g. Technological Characteristics
The previously cleared NM3 monitor contains a flow and carbon dioxide monitoring component consisting of the previously 510(k) cleared Mercury module with Capnostat 5, a pulse oximetry component consisting of the previously 510(k) cleared Masimo Rainbow Set OEM module, and a cardiac output monitoring component. The Mercury module with Capnostat 5 provides continuous monitoring of respiratory flow and pressure, and CO2. Additionally, besides the Capnostat 5 sensor, the previously cleared LoFlo C5 sidestream module may be interfaced to the Mercury Module. The Masimo Rainbow Set OEM module provides continuous monitoring of SpO2 and pulse rate. The cardiac output monitoring component provides cardiac output measurement capabilities using the partial CO2 rebreathing method. The VentAssist software option provides additional screens to display parameters in the predicate Philips NM3 Monitor in a bar chart format, and adds a new work of breathing (WOB/min) parameter computed with a fixed Artificial Neural Net (ANN). A new user-selectable soft key is available to receive on-demand advice. The advice is determined using a fuzzy-logic rule based system which provides on-demand advice for the setting of the PSV level and ventilator support levels based upon WOB/min, breathing frequency, tidal volume, ideal body weight, and end-tidal CO2. The advice is based upon a set of fuzzy logic rules developed and refined in conjunction with experienced critical care clinicians. The rule set is based upon commonly used measures of the patient's respiratory condition including patient effort and ventilation. The predicated Dräger Medical SmartCare™/PS software with the exception of work of breathing/min uses the same measures (i.e. breathing rate, tidal volume and end tidal CO2)
### h. Validation
.
The validation of the noninvasive work of breathing per minute parameter consisted of (a) validating the front end for esophageal pressure measurement with a cleared device, Bicore CP-100 Cardiopulmonary Monitor, (b) comparing invasive measurements of WOB/min using the esophageal balloon to noninvasive measures of WOB/min derived from proximal flow and airway pressure data collected from ventilated adults in the ICU
{3}------------------------------------------------
(MICU, CICU, SICU, burn unit, step-down unit) who were breathing spontaneously on pressure support ventilation (bias and precision of 0.84 ±2.2 J/min) and (c) validating that the WOB/min algorithm was properly implemented in the embedded system, the Philips NM3 Monitor.
The rule set as implemented in the PS/V Advisor was tested in two separate clinical studies. One study tested if clinicians agreed with the PS Advisor recommendations, and the second study was similar but done blinded so the clinician was unable to see the recommendations of the PS Advisor. A peer-reviewed paper (Banner et al, Chest, 133(3):697-703, 2008) summarizing the results of the first study was published in 2008.
We believe the bench and clinical testing (discussed above) has demonstrated that the VentAssist PS/V advisor is substantially equivalent to the predicate devices cited in section d.
{4}------------------------------------------------
Image /page/4/Picture/1 description: The image shows the logo for the U.S. Department of Health & Human Services. The logo consists of a stylized eagle or bird-like symbol with three wing-like shapes, representing the department's mission to protect the health of all Americans. The logo is surrounded by a circular border containing the text "DEPARTMENT OF HEALTH & HUMAN SERVICES USA".
Food and Drug Administration 10903 New Hampshire Avenue Document Control Room -WO66-G609 Silver Spring, MD 20993-0002
Mr. Kevin Mader Manager of Quality Assurance and Regulatory Affairs Respironics Novameterix, LLC Critical Care 5 Technology Drive Wallingford, Connecticut 06492
JUN - 2 2011
Re: K103578
Trade/Device Name: Philips NM3 Monitor with VentAssist, Model 7900 Regulation Number: 21 CFR 868.1850 Regulation Name: Monitoring Spirometer Regulatory Class: II Product Code: BZK Dated: May 31, 2011 Received: June 1, 2011
Dear Mr. Mader:
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, market 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 (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.
{5}------------------------------------------------
### Page 2- Mader
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 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 (OS) 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/cdrh/mdr/ 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 (240) 276-3150 or at its Internet address http://www.fda.gov/cdrh/industry/support/index.html.
Sincerely yours.
Nh for
Anthony D. Watson, B.S., M.S., M.B.A. Director Division of Anesthesiology, General Hospital, Infection Control and Dental Devices Office of Device Evaluation Center for Devices and Radiological Health
{6}------------------------------------------------
## Indications for Use
# 510(k) Number (if known): 长l03578
Device Name: Philips NM3 Respiratory Profile Monitor with VentAssist, Model 7900
Indications for Use:
The intended use of the Philips NM3 Respiratory Profile Monitor, Model 7900, is to provide:
- cardiac output monitoring via the method of partial rebreathing in adult patients receiving . mechanical ventilation during general anesthesia and in the intensive care unit (ICU).
- spirometric, and carbon dioxide monitoring in neonatal, pediatric and adult patients during . general anesthesia and in the intensive care unit (ICU) and the emergency department Separate combination CO2/flow sensors are provided for adult, pediatric and (ED). neonatal use.
- continuous, non-invasive monitoring of functional arterial oxygen saturation and pulse . rate in neonatal, pediatric and adult patients during both no motion and motion conditions and for patients who are well or poorly perfused during general anesthesia and in the intensive care unit (ICU) and the emergency department (ED).
The intended use of the VentAssist software option is to provide:
- non-invasive monitoring of work of breathing per minute in adult patients receiving . pressure support mechanical ventilation.
- on-demand advice in mechanically ventilated adult patients as prescribed by the ● caregiver regarding (a) modifications to the current pressure support settings in order to assess the work of breathing and breathing pattern and (b) modifications to ventilation in order to maintain end-tidal CO2 in a range determined by the physician. The patients should be hemodynamically stable and must be breathing spontaneously.
Prescription Use (Part 21 CFR 801 Subpart D)
AND/OR
Over-The-Counter Use (21 CFR 801 Subpart C)
## (PLEASE DO NOT WRITE BELOW THIS LINE-CONTINUE ON ANOTHER PAGE IF NEEDED)
Concurrence of CDRH, Office of Device Evaluation (ODE)
L. Schuttler
(Division Sign-Un) Division of Anesthesiology, General Hospital Infection Control, Dental Devices
n 103578 510(k) Number:
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.