K250912 · Cytovale, Inc. · QUT · Nov 19, 2025 · Microbiology
Device Facts
Record ID
K250912
Device Name
IntelliSep Test
Applicant
Cytovale, Inc.
Product Code
QUT · Microbiology
Decision Date
Nov 19, 2025
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.3215
Device Class
Class 2
Attributes
Real-World Evidence
Real-World Evidence
Submission
Device
Sponsor
RWD Sources
RWE Use Summary
Key Tags
K250912 · Nov 19, 2025
IntelliSep Test
Cytovale, Inc.
Electronic Health Record (EHR); Retrospective physician clinical adjudication
Retrospective clinical adjudication of sepsis status based on EHR data was used to compare the performance of the modified device cartridge against the predicate device.
The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample. The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use.
Device Story
The IntelliSep test is an in vitro diagnostic assay used in the Emergency Department to assess immune activation in patients with suspected infection. The device processes 100 μL of K2 EDTA venous whole blood; the Cytovale System instrument automatically lyses erythrocytes and washes leukocytes. The prepared sample is loaded into a microfluidic cartridge for deformability cytometry, which measures the biophysical properties of individual leukocytes. The system calculates an IntelliSep Index (ISI) ranging from 0.1 to 10.0, categorized into three interpretation bands of increasing sepsis likelihood. Clinicians use these results alongside other clinical information to aid in the early detection of sepsis with organ dysfunction. The device does not identify causative pathogens and is not a standalone diagnostic tool.
Clinical Evidence
Bench testing only. Comparison of ISI results between original and modified cartridges using identical patient samples demonstrated no performance differences. Assay precision and reproducibility were evaluated, confirming performance equivalence to K220991. Shipping validation performed per ASTM D4169 and ISTA 7D confirmed package integrity and consistent device performance after conditioning.
Indicated for adult patients with signs and symptoms of infection presenting to the Emergency Department to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing.
Regulatory Classification
Identification
A device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis is identified as an in vitro device intended for the detection and qualitative and/or quantitative measurement of one or more non-microbial analytes in human clinical specimens to aid in the assessment of patients with suspected sepsis when used in conjunction with clinical signs and symptoms and other clinical and laboratory findings.
Special Controls
A device to detect and measure non-microbial analyte(s) in human clinical specimens to aid in assessment of patients with suspected sepsis must comply with the following special controls:
*Classification.* Class II (special controls). The special controls for this device are:(1) Premarket notification submissions must include the device's detailed Indications for Use statement describing what the device detects and measures, the results provided to the user, whether the measure is qualitative and/or quantitative, the clinical indications for which the test is to be used, and the specific population(s) for which the device use is intended.
(2) Premarket notification submissions must include detailed documentation of the device description, including (as applicable), all device components, software, ancillary reagents required but not provided, explanation of the device principle and methodology, and for molecular devices include detailed documentation of the primer/probe sequence, design, and rationale for sequence selection.
(3) Premarket notification submissions must include detailed documentation of applicable analytical studies, such as, analytical sensitivity (Limit of Detection, Limit of Blank, and Limit of Quantitation), precision, reproducibility, analytical measuring range, interference, cross-reactivity, and specimen stability.
(4) Premarket notification submissions must include detailed documentation of a prospective clinical study or, if appropriate, results from an equivalent sample set. This detailed documentation must include the following information:
(i) Results must demonstrate adequate device performance relative to a well-accepted comparator.
(ii) Clinical sample results must demonstrate consistency of device output throughout the device measuring range likely to be encountered in the Intended Use population.
(iii) Clinical study documentation must include the original study protocol (including predefined statistical analysis plan), study report documenting support for the Indications for Use(s), and results of all statistical analyses.
(5) Premarket notification submissions must include evaluation of the level of the non-microbial analyte in asymptomatic patients with demographic characteristics (
*e.g.,* age, racial, ethnic, and gender distribution) similar to the Intended Use population.(6) As part of the risk management activities performed under 21 CFR 820.10(c) design and development, you must document an appropriate end user device training program that will be offered as part of your efforts to mitigate the risk of failure to correctly operate the instrument.
(7) A detailed explanation of the interpretation of results and acceptance criteria must be included in the device's 21 CFR 809.10(b)(9) compliant labeling, and a detailed explanation of the interpretation of the limitations of the samples (
*e.g.,* collected on day of diagnosis) must be included in the device's 21 CFR 809.10(b)(10) compliant labeling.
{0}
FDA
U.S. FOOD & DRUG
ADMINISTRATION
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
ASSAY AND INSTRUMENT
## I Background Information:
A 510(k) Number
K250912
B Applicant
Cytovale, Inc.
C Proprietary and Established Names
IntelliSep test
D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| QUT | Class II | 21 CFR 866.3215 - Device To Detect And Measure Non-Microbial Analyte(S) In Human Clinical Specimens To Aid In Assessment Of Patients With Suspected Sepsis | MI - Microbiology |
## II Submission/Device Overview:
A Purpose for Submission:
To obtain a substantial equivalence determination for modifications to the Cytovale IntelliSep test exterior cartridge housing, inlet filter assembly, and cartridge primary packaging.
B Measurand:
White blood cells in K2 EDTA venous whole blood
C Type of Test:
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
Semi-quantitative in vitro assay that measures deformability cytometry of leukocytes
## Intended Use/Indications for Use:
### A Intended Use(s):
See Indications for Use below.
### B Indication(s) for Use:
The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample.
The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use.
### C Special Conditions for Use Statement(s):
Rx - For Prescription Use Only
### D Special Instrument Requirements:
The IntelliSep test is intended for use on the Cytovale System instrument.
## Device/System Characteristics:
### A Device Description:
The Cytovale IntelliSep test is an in vitro clinical test to aid in the early detection of sepsis with organ dysfunction manifesting within the first three days after testing. It assesses the state of immune activation in patients with clinical suspicion of infection who present in the Emergency Department (ED). The IntelliSep test and Cytovale System instrument were previously cleared in K250513. Refer to the published decision summary for additional information. The purpose of these 510(k) submissions was to validate modifications to the Cytovale IntelliSep test exterior cartridge housing, inlet filter assembly, and cartridge primary packaging.
### B Principle of Operation:
K250912 - Page 2 of 10
{2}
Laboratory operators transfer 100 μL of whole blood into a sample preparation tube. The operator inserts the sample preparation tube into the Cytovale System instrument, which automatically lyses erythrocytes and washes purified leukocytes in a diluent producing a total volume of approximately 1 mL of prepared sample. The operator transfers the prepared sample to the IntelliSep test cartridge for imaging and analysis on the Cytovale System.
The biophysical properties of individual leukocytes are measured using a microfluidic deformability cytometry technique. Based on these measurements, the test assigns a numerical value, the IntelliSep Index, ranging from 0.1-10.0. Results are stratified into three discrete interpretation bands (Band 1, Band 2, Band 3) of increasing sepsis likelihood.
## C Instrument Description Information:
1. Instrument Name:
Cytovale System
2. Specimen Identification:
The IntelliSep test on the Cytovale System is validated for use only with K2 EDTA anticoagulated whole blood.
3. Specimen Sampling and Handling:
The Sample Processing Module of the Cytovale System receives 100 μL of K2 EDTA anticoagulated whole blood in the Sample Preparation Tube and processes the specimen into a white blood cell sample suspended in Cytovale Diluent. The operator injects 1.0 mL of the prepared sample into the IntelliSep test cartridge inlet well and inserts the cartridge into the Cell Imaging Module for analysis. Once the analysis is complete, the Cartridge is disposed of according to institutional guidelines on biohazardous waste.
4. Calibration:
All necessary configuration, qualification, and calibration (where applicable) is performed by a Cytovale technician during installation. Further configuration and calibration procedures are not required by the operator.
Cytovale recommends that the System be serviced every three months of use, based on the initial installation date (or based on the previous service date), or every 3500 tests, whichever comes first.
5. Quality Control:
The IntelliSep test on the Cytovale System has a two-level quality control (i.e., L1 and L2) set called the IntelliSep Quality Control Kit, derived from stabilized whole blood. The kit contains assayed controls with values within the established measuring range. This test-specific Quality Control Kit is tested the same as patient specimens during processing. The mean and range of assay values for a particular lot of the IntelliSep Quality Control Kit (L1 and L2) are reported on the IntelliSep Quality Control Assay Sheet. An Assay Sheet is
K250912 - Page 3 of 10
{3}
provided with each lot of materials. They are lot specific and are derived from replicate analyses on Cytovale Systems.
External controls should be tested in accordance with laboratory practices and accreditation requirements per 42 CFR 493.1256.
## V Substantial Equivalence Information:
### A Predicate Device Name(s):
IntelliSep test
### B Predicate 510(k) Number(s):
K220991
### C Comparison with Predicate(s):
| Device & Predicate Device(s): | K250912 | K220991 |
| --- | --- | --- |
| Device Trade Name | IntelliSep test | IntelliSep test |
| General Device Characteristic Similarities | | |
| Intended Use/Indications For Use | The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample.
The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of | The Cytovale IntelliSep test is a semi-quantitative test that assesses cellular host response via deformability cytometry of leukocyte biophysical properties and is intended for use in conjunction with clinical assessments and laboratory findings to aid in the early detection of sepsis with organ dysfunction manifesting within the first 3 days after testing. It is indicated for use in adult patients with signs and symptoms of infection who present to the Emergency Department. The test is performed on an EDTA anticoagulated whole blood sample.
The IntelliSep test generates an IntelliSep Index value that falls within one of three discrete interpretation bands based on the probability of sepsis with organ dysfunction manifesting within the first three days after testing. The IntelliSep test represents the probability of the clinical syndrome of sepsis and is intended to be used alongside other clinical information and clinical judgement. It does not identify the causative agent of |
K250912 - Page 4 of 10
{4}
K250912 - Page 5 of 10
| | infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use. | agent of infection and should not be used as the sole basis to determine the presence of sepsis. The IntelliSep test is intended for in vitro diagnostic use. |
| --- | --- | --- |
| Assay Method | Microfluidic deformability cytometry | same |
| Analyte | Leukocyte biophysical properties | same |
| Specimen Type | Human venous whole blood (K2 EDTA) | same |
| Instrument Platform | Cytovale System | same |
| Result Output | IntelliSep Index | same |
| **General Device Characteristic Differences** | | |
| Filter Material | proprietary synthetic polyamide | nylon |
| Filter Size | 10 μm | 25 μm |
| Cartridge Material | acrylic | cyclo olefin polymer |
| Cartridge Cover Material | cyclo olefin polymer | acrylic |
| Packaging | thermoformed sealed pouch | thermoformed clamshell |
VI Standards/Guidance Documents Referenced:
- ISO 14971:2019 Medical devices - Application of risk management to medical devices
- ISO TR24971:2013 Medical devices – Guidance on the application of 14971
- ISO 1348:2016 Medical devices – Quality Management Systems – Requirements for regulatory purposes
VII Performance Characteristics (if/when applicable):
A Analytical Performance:
1. Precision/Reproducibility:
Within-Laboratory Precision with Donor Samples
A study was conducted to evaluate the within-laboratory precision for the modified assay cartridges. Specifically, five donor samples were tested in triplicate by two operators on two instruments for two separate runs spaced at least one hour apart for a total of twenty-four tests per donor. For each sample, one lot of IntelliSep test cartridges, one lot of Cytovale Reagent Kit, one lot of Cytovale Diluent, and one lot of Cytovale Cleanse was used.
A total of 123 assays were performed for the study with four run failures encountered (3.3%). Three run failures were clog errors on the same instrument and Band 3 sample combination. One error was caused by a facility power outage. Due to the power outage, no ISI was reported for this assay. The assay was successfully rerun within the study design and time parameters. Table 1 below, summarizes the within-laboratory precision study results.
{5}
Table 1: Within-Laboratory Precision Results
| Sample | N | Mean | Between-operator (SD) | Between-instrument (SD) | Between-run (SD) | Repeatability (SD) | Within-lab precision (SD) |
| --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | 24 | 6.37 | 0.00 | 0.00 | 0.03 | 0.22 | 0.22 |
| 2 | 24 | 3.84 | 0.00 | 0.13 | 0.10 | 0.16 | 0.23 |
| 3 | 24 | 7.29 | 0.06 | 0.18 | 0.30 | 0.44 | 0.56 |
| 4 | 23 | 7.37 | 0.00 | 0.42 | 0.00 | 0.49 | 0.65 |
| 5 | 24 | 5.53 | 0.00 | 0.21 | 0.00 | 0.22 | 0.30 |
| Total | 119 | - | 0.00 | 0.04 | 0.09 | 0.38 | 0.39 |
The acceptance criterion for within-laboratory precision is standard deviation $\leq 1.0$ units of IntelliSep Index. The study design and results from within-laboratory precision testing are acceptable.
# Reproducibility with Quality Control Materials
Since samples must be tested immediately after collection on the IntelliSep test, a site-to-site reproducibility study was conducted with IntelliSep Quality Control materials on the Cytovale System. In summary, five replicates of two Quality Control level materials (Level 1 and 2) were tested over five days at two independent sites for a total of 25 valid test results per control at each site. A single lot of modified IntelliSep test cartridges was used across all test sites and replicates. Table 2 summarizes the reproducibility study results at each site and both sites combined.
Table 2: Reproducibility Results
| Sample | Site | Mean ISI | N | SD |
| --- | --- | --- | --- | --- |
| QC Level 1 | 1 | 3.63 | 25 | 0.39 |
| QC Level 2 | | 7.87 | 25 | 0.37 |
| QC Level 1 | 2 | 3.16 | 25 | 0.43 |
| QC Level 2 | | 7.27 | 25 | 0.56 |
| QC Level 1 | Both Sites | 3.4 | 50 | 0.47 |
| QC Level 2 | | 7.57 | 50 | 0.56 |
The study acceptance criteria was a reproducibility standard deviation of 1.5 ISI values. The estimated site-to-site reproducibility for the modified cartridges is acceptable.
2. Linearity:
Not applicable.
3. Analytical Specificity/Interference:
Not applicable.
4. Assay Reportable Range:
K250912 - Page 6 of 10
{6}
The measuring range of the IntelliSep test is between 0.1 and 10.0 units, divided into three interpretation bands: Band 1 (ISI = 0.1-4.9), Band 2 (ISI = 5.0-6.2), and Band 3 (6.3-10.0).
5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
There were no changes that affect stability claims or calibration of the device. For additional information, refer to the published decision summary for K220991.
6. Detection Limit:
Not applicable.
7. Assay Cut-Off:
There were no changes to the assay cutoff. Refer to the published decision summary for K220991 for additional information.
8. Accuracy (Instrument):
Not applicable.
9. Carry-Over:
A carry-over study was conducted to support the original clearance of the device. Refer to the published decision summary for K220991 for additional information.
## B Comparison Studies:
Clinical validation of the Cytovale IntelliSep test in the intended use population was previously established in K220991. Refer to the published decision summary for additional information. The purpose of these 510(k) submissions is to establish equivalence between the modified Cytovale IntelliSep test cartridge and the originally cleared predicate device.
1. Method Comparison with Predicate Device:
A method comparison study was performed with 109 donor samples, of which 51 were adjudicated septic and 57 were adjudicated as not septic. For each donor sample, two replicates were performed on one cleared predicate cartridge lot and one modified cartridge lot for a total of four tests per donor and 432 total valid tests.
Weighted Deming regression analysis was performed on the dataset to assess the equivalence between the two cartridges. Regression analysis was used to calculate systematic bias at the two medical decision boundaries, corresponding to X=5.0 ISI and X=6.2 ISI. Equivalence was assessed at each boundary using two one-sided tests conducted at α = 0.05. Results from this study are summarized in Table 3 and Table 4 below.
K250912 - Page 7 of 10
{7}
The acceptance criterion for this study was that equivalence tests pass at the 95% confidence level for each medical decision point. Table 3 below summarizes the Deming regression results.
Table 3: Medical decision boundary assessment with 95% Confidence Interval for weighted Deming regression using the entire data set.
| Medical Decision Boundary | Regression Value [95% CI] (test cartridge ISI) | Acceptance Bounds |
| --- | --- | --- |
| 5.0 | 5.3 [5.2, 5.5] | [4.5, 5.5] |
| 6.2 | 6.5 [6.4, 6.6] | [5.7, 6.7] |
Table 4 summarizes the comparison between the cleared predicate device results and the modified IntelliSep test results by bin including overall agreement.
Table 4: Modified IntelliSep test Bin Results – Comparison to the Predicate Device
| Modified IntelliSep test | Cleared Predicate Device | | |
| --- | --- | --- | --- |
| | Bin 1 | Bin 2 | Bin 3 |
| Bin 1 | 20 | 0 | 0 |
| Bin 2 | 5 | 26 | 0 |
| Bin 3 | 0 | 8 | 49 |
| Agreement | 80.00% | 76.47% | 100.00% |
| | (20/25) | (26/34) | (49/49) |
No clinical samples evaluated shifted to a non-adjacent reporting bin when measured by the modified IntelliSep test. Cumulatively, the data summarized establishes equivalent performance between the modified IntelliSep test and the cleared predicate device.
Clinical data and adjudication results were collected from the patient Electronic Health Record (EHR). A structured adjudication process was used to classify the presence or absence of sepsis by Sepsis-3 criteria: To be considered positive for Sepsis-3, patients must have each of three components: (1) infection (present on presentation to the ED), (2) organ dysfunction (manifesting within 3 days of the ED visit), and (3) organ dysfunction caused by a dysregulated host response to the infection.
The Site Principal Investigator, with access to the medical records but blinded to study ISI values, performed a retrospective clinical adjudication by completing a structured adjudication process in the REDCap system, recording pertinent clinical information about the determination. Table 5 summarizes the method comparison study results when comparing IntelliSep test modified cartridge results to a physician adjudication of sepsis.
Table 5: Modified IntelliSep test – Comparison to Sepsis Adjudication
K250912 - Page 8 of 10
{8}
| IntelliSep Result (Modified Cartridge) | Adjudicated Septic (+) | Adjudicated Not-Septic (-) | Sepsis Predictive Value (80% CI) | Sepsis Likelihood Ratio |
| --- | --- | --- | --- | --- |
| Band 1 | 1 | 19 | 5.0% (0.5%, 18.1%) | 0.06 |
| Band 2 | 13 | 18 | 41.9% (29.7%, 55.0%) | 0.81 |
| Band 3 | 37 | 20 | 64.9% (55.6%, 73.4%) | 2.07 |
| | | | Sepsis Prevalence | |
| Total | 51 | 57 | 47.20% | N/A |
Table 6 summarizes the method comparison study results when comparing predicate device results to a physician adjudication of sepsis.
Table 6: IntelliSep Tests Original and Modified Cartridge - Sepsis Predictive Values and Sepsis Likelihood Ratios
| | Modified Cartridge | | Original Cartridge | |
| --- | --- | --- | --- | --- |
| IntelliSep Result | Sepsis Predictive Value (80% CI) | Sepsis Likelihood Ratio | Sepsis Predictive Value (80% CI) | Sepsis Likelihood Ratio |
| Band 1 | 5.0% (0.5%, 18.1%) | 0.06 | 8.0% (2.1%, 19.9%) | 0.10 |
| Band 2 | 41.9% (29.7%, 55.0%) | 0.81 | 47.1% (35.0%, 59.4%) | 0.99 |
| Band 3 | 64.9% (55.6%, 73.4%) | 2.07 | 67.3% (57.3%, 76.3%) | 2.07 |
The predictive value of the ISI band results for identifying patients with sepsis in the modified assay cartridge increased with increasing score value with non-overlapping confidence intervals for non-adjacent bands. The results for the modified assay cartridge were also similar to results obtained from the original unmodified version. These data support equivalent performance of the modified IntelliSep test cartridge.
2. Matrix Comparison:
Not applicable.
C Clinical Studies:
1. Clinical Sensitivity:
The clinical performance in the intended use population was established in the original $510(\mathrm{k})$ for the device. Refer to the published decision summary for K220991 for additional information.
2. Clinical Specificity:
The clinical performance in the intended use population was established in the original $510(\mathrm{k})$ for the device. Refer to the published decision summary for K220991 for additional information.
3. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable):
K250912 - Page 9 of 10
{9}
Not applicable.
## D Clinical Cut-Off:
There were no changes to the assay cut-offs or bin interpretations. Refer to the published decision summary for K220991 for additional information.
## E Expected Values/Reference Range:
The expected values and reference ranges for the assay were established in the original 510(k) submission for the device. Refer to the published decision summary for K220991 for additional information.
## F Other Supportive Instrument Performance Characteristics Data:
Not applicable.
## VIII Proposed Labeling:
The labeling supports the finding of substantial equivalence for this device.
## IX Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
K250912 - Page 10 of 10
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.
