Cytomics FC 500 Series (MPL or MCL) Flow Cytometer
Applicant
Beckman Coulter
Product Code
GKZ · Hematology
Decision Date
Sep 6, 2019
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 864.5220
Device Class
Class 2
Attributes
Real-World Evidence
Real-World Evidence
Submission
Device
Sponsor
RWD Sources
RWE Use Summary
Key Tags
K182886 · Sep 6, 2019
Cytomics FC 500 Series (MPL or MCL) Flow Cytometer
Beckman Coulter
Customer complaint casefiles; Clinical site listmode datafiles (LMD) from routine laboratory use
Real-world data files were used to evaluate the performance (False Negative and False Positive rates) of the Software Detection Tool (SDT) in detecting data compromised by actual amp board failures in clinical laboratories.
Real-world data; Customer complaints; Clinical site data; Software performance validation
Clinical Evidence
Study Design
Population
Comparator
Key Endpoints
Real World Data Performance Evaluation; Retrospective analysis of clinical and customer-sourced datafiles; Follow-up/Duration: Not applicable
Clinical laboratory samples processed on instruments with known or suspected amp board failures; Sample Size: 409 datafiles; Number of Sites: Not specified (includes multiple customer sites and one clinical site)
Not applicable for this study
False Negative Rate (FNR) and False Positive Rate (FPR) of the Software Detection Tool
Indications for Use
The Cytomics FC 500 MPL is a system for the qualitative and quantitative measurement of biological and physical properties of cells and other particles. These properties are measured when the cells pass through one or two laser beams in single-file. The tetraCXP SYSTEM for Cytomics FC 500 flow cytometry systems is an automated analysis method for simultaneous identification and enumeration of lymphocyte subpopulations (CD3+, CD4+, CD8+, CD19+ and CD56+) combining four-color fluorescent monoclonal antibody reagents, quality control reagents, optional absolute count reagent and CXP software. The systems with CYTO-STAT tetraCHROME CD45-FITC/CD4-PD/CD8-ECD/CD3-PC5 Monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of Total CD3+ (T cells), Total CD4+, Total CD8+, Dual CD3+/CD4+, Dual CD3+/CD8+ lymphocyte percentages and absolute counts as well as the CD4/CD8 ratio in whole blood flow cytometry. The systems with CD45-FITC/CD56-PC/CD19-ECD/CD3-PC5, the total lymphocyte percentage can be obtained. CD45-FITC/CD56-PE/CD19-ECD/CD3-PC5 monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of total CD19+ (B cells) and CD3−/CD56+ (NK cells) lymphocyte percentages and absolute counts in whole blood flow cytometry. The total lymphocyte percentage can be obtained as well.
Device Story
Flow cytometer uses hydrodynamic focusing to align cells in single-file through laser beams (488 nm and 635/633 nm). Measures forward scatter (size), side scatter (granularity), and fluorescence (surface antigens) via seven sensors. Converts light to voltage pulse signals; workstation software (MXP/CXP) processes signals to generate data plots, statistics, and reports. Corrective software update adds Software Detection Tool (SDT) to mitigate signal loss/drift. SDT monitors data acquisition; alerts operator if signal failure detected; aborts run or flags data for visual review. Used by laboratory technicians for clinical diagnostics. Output enables identification/enumeration of lymphocyte subsets, aiding clinical decision-making regarding immune status.
Clinical Evidence
No clinical trials; evidence based on analytical performance testing. SDT performance evaluated via digital, electronic, and physical simulations (n=2344-2726 per panel) and real-world data (n=409). Pre-acquisition detection sensitivity was 100%. Post-acquisition False Negative Rates (FNR) ranged from 0.04% to 2.79% across panels. A reader study (n=100 files) with Tier 1 and Tier 2 (pathologist) review demonstrated an FNR of 1.0% (95% CI: 0.2-5.5%) for manual review of flagged data, confirming the effectiveness of the mitigation strategy.
Technological Characteristics
Flow cytometer with 488nm laser excitation; 7 detector channels (2 APD for scatter, 5 PMT for fluorescence). Sample handling via carousel (MCL) or plate-loader (MPL). Software architecture includes CXP/MXP/BCAP modules. SDT algorithm integrated into software to monitor amp board delay lines. Connectivity via workstation; no wireless transmission. Sterilization not applicable (instrument).
Indications for Use
Indicated for qualitative and quantitative measurement of biological/physical properties of cells/particles in whole blood. Used for identification and enumeration of lymphocyte subpopulations (CD3+, CD4+, CD8+, CD19+, CD56+) and CD4/CD8 ratios. Intended for prescription use in clinical laboratory settings.
Regulatory Classification
Identification
An automated differential cell counter is a device used to identify one or more of the formed elements of the blood. The device may also have the capability to flag, count, or classify immature or abnormal hematopoietic cells of the blood, bone marrow, or other body fluids. These devices may combine an electronic particle counting method, optical method, or a flow cytometric method utilizing monoclonal CD (cluster designation) markers. The device includes accessory CD markers.
Special Controls
*Classification.* Class II (special controls). The special control for this device is the FDA document entitled “Class II Special Controls Guidance Document: Premarket Notifications for Automated Differential Cell Counters for Immature or Abnormal Blood Cells; Final Guidance for Industry and FDA.”
{0}
K182286
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY
A. 510(k) Number:
K182886
B. Purpose for Submission:
Instrument modification
C. Manufacturer and Instrument Name:
Beckman Coulter, Inc.
Cytomics FC500 MCL and MPL Flow Cytometers
D. Type of Test or Tests Performed:
Flow cytometric immunoassays: cell identification with quantitative cell counts
E. System Descriptions:
1. Device Description:
The Cytomics FC500 MCL and MPL Flow Cytometers are instruments for the qualitative and quantitative measurement of biological and physical properties of cells and other particles. These properties are measured when the cells pass through one or two laser beams in single-file. The MCL configuration denotes a carousel-loader unit for sample tubes; the MPL configuration denotes a plate-loader unit, in addition to sample tubes.
These instruments include a 488nm excitation laser, and an optional 633nm solid-state or 635nm HeNe laser. IVD assays cleared for use on the FC500 cytometers employ the 488nm laser for excitation, no assays that employ the 633/635 lasers have been cleared by the FDA and clearance of those lasers is not the subject of this 510(k). All configurations of the FC500 cytometer possess a total of seven detector channels – two avalanche photodiodes (APD), for forward- and side-scatter; and five photomultiplier tubes (PMT, FL1–FL5) for detection of specific fluorescence. Seven amp boards (one for each detector circuit) amplify and synchronize analog electronic signals from photodetectors, prior to conversion to digital signals.
The software architecture for the FC500 instrument-controlling workstation includes CXP for acquisition and analysis of closed-system IVD assays, MXP for more advanced acquisition and analysis including LDT assays, and BCAP for necessary user and account permissions controls for review of quarantined files.
This 510(k) submission concerns software update patches (CXP v2.3, MXP v2.3),
1
{1}
enabling detection of delay line failures in the amp board for each detector channel, as described in RCL181101. As a corrective Change Being Effected (CBE) response, this software update contains three algorithmic processes as part of a Software Detection Tool (SDT) for:
1) electronic detection of pre-acquisition signal failures in amp board delay lines
2) mid- or post-acquisition detection of signal failures by automated digital analysis of parameter-by-time signal for each detection channel
3) predictive quarantine of datafiles suspected of being compromised, due to signal failure, for operator review prior to release or dismissal
2. Principles of Operation:
The Cytomics FC500 cytometric instrument contains a fluidics system designed to pass cells or particles through one or more laser beams in succession. Scattered and fluoresced light from these events is collected by APD and PMT detectors, converted to digital signal, and collected by the instrument workstation.
3. Modes of Operation:
Does the applicant's device contain the ability to transmit data to a computer, webserver, or mobile device?
Yes ☐ X ☐ or No ☐
Does the applicant's device transmit data to a computer, webserver, or mobile device using wireless transmission?
Yes ☐ or No ☐ X ☐
4. Specimen Identification:
Barcode reader or manual entry
5. Specimen Sampling and Handling:
MCL: 32 (12×72 mm) tubes
MPL: 24- or 96-well plates, or 12×72mm tubes
6. Calibration:
Assay-dependent
7. Quality Control:
Assay-dependent
K182286
{2}
8. Software:
FDA has reviewed applicant's Hazard Analysis and Software Development processes for this line of product types:
Yes ☐ X ☐ or No ☐
F. Regulatory Information:
1. Regulation section:
21 CFR §864.5220 – Automated Differential Cell Counter
2. Classification:
Class II
3. Product code:
GKZ: Automated Differential Cell Counter
4. Panel:
Hematology (81)
G. Intended Use:
1. Indication(s) for Use:
K071681 (MPL)
The Cytomics FC 500 MPL is a system for the qualitative and quantitative measurement of biological and physical properties of cells and other particles. These properties are measured when the cells pass through one or two laser beams in single-file.
K030828 (MCL)
The tetraCXP SYSTEM for Cytomics FC 500 flow cytometry systems is an automated analysis method for simultaneous identification and enumeration of lymphocyte subpopulations (CD3+, CD4+, CD8+, CD19+ and CD56+) combining four-color fluorescent monoclonal antibody reagents, quality control reagents, optional absolute count reagent and CXP software. The systems with CYTO-STAT tetraCHROME CD45-FITC/CD4-PD/CD8-ECD/CD3-PC5 Monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of Total CD3+ (T cells), Total CD4+, Total CD8+, Dual CD3+/CD4+,
K182286
{3}
Dual CD3+/CD8+ lymphocyte percentages and absolute counts as well as the CD4/CD8 ratio in whole blood flow cytometry. The systems with CD45-FITC/CD56-PC/CD19-ECD/CD3-PC5, the total lymphocyte percentage can be obtained. CD45-FITC/CD56-PE/CD19-ECD/CD3-PC5 monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of total CD19+ (B cells) and CD3−/CD56+ (NK cells) lymphocyte percentages and absolute counts in whole blood flow cytometry. The total lymphocyte percentage can be obtained as well.
2. Special Conditions for Use Statement(s):
For In Vitro Diagnostic Use
H. Substantial Equivalence Information:
1. Predicate Device Name(s) and 510(k) numbers:
K030828: TetraCXP System (MCL)
K071681: Cytomics FC500 MPL Flow Cytometer
2. Comparison with Predicate Device:
| Similarities | | | |
| --- | --- | --- | --- |
| Item | Device | Predicate (K030828) | Predicate (K071681) |
| Intended Use | The Cytomics FC 500 MPL is a system for the qualitative and quantitative measurement of biological and physical properties of cells and other particles. These properties are measured when the cells pass through one or two laser beams in single-file. The tetraCXP SYSTEM for Cytomics FC 500 flow cytometry systems is an automated analysis method for simultaneous identification and enumeration of lymphocyte subpopulations (CD3+, CD4+, CD8+, CD19+ and CD56+) combining four-color fluorescent monoclonal antibody reagents, quality control reagents, optional absolute count reagent and CXP software. The systems with CYTO-STAT tetraCHROME CD45-FITC/CD4-PD/CD8-ECD/CD3-PC5 Monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", | The tetraCXP SYSTEM for Cytomics FC 500 flow cytometry systems is an automated analysis method for simultaneous identification and enumeration of lymphocyte subpopulations (CD3+, CD4+, CD8+, CD19+ and CD56+) combining four-color fluorescent monoclonal antibody reagents, quality control reagents, optional absolute count reagent and CXP software. The systems with CYTO-STAT tetraCHROME CD45-FITC/CD4-PD/CD8-ECD/CD3-PC5 Monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", | The Cytomics FC 500 MPL is a system for the qualitative and quantitative measurement of biological and physical properties of cells and other particles. These properties are measured when the cells pass through one or two laser beams in single-file. |
K182286
{4}
| Similarities | | | |
| --- | --- | --- | --- |
| Item | Device | Predicate (K030828) | Predicate (K071681) |
| | control reagents, optional absolute count reagent and CXP software. The systems with CYTO-STAT tetraCHROME CD45-FITC/CD4-PD/CD8-ECD/CD3-PC5 Monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of Total CD3+ (T cells), Total CD4+, Total CD8+, Dual CD3+/CD4+, Dual CD3+/CD8+ lymphocyte percentages and absolute counts as well as the CD4/CD8 ratio in whole blood flow cytometry. The systems with CD45-FITC/CD56-PC/CD19-ECD/CD3-PC5, the total lymphocyte percentage can be obtained. CD45-FITC/CD56-PE/CD19-ECD/CD3-PC5 monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of total CD19+ (B cells) and CD3-/CD56+ (NK cells) lymphocyte percentages and absolute counts in whole blood flow cytometry. The total lymphocyte percentage can obtained as well. | allowing the identification and enumeration of Total CD3+ (T cells), Total CD4+, Total CD8+, Dual CD3+/CD4+, Dual CD3+/CD8+ lymphocyte percentages and absolute counts as well as the CD4/CD8 ratio in whole blood flow cytometry. The systems with CD45-FITC/CD56-PC/CD19-ECD/CD3-PC5, the total lymphocyte percentage can be obtained. CD45-FITC/CD56-PE/CD19-ECD/CD3-PC5 monoclonal antibody reagent is intended "For In Vitro Diagnostic Use", allowing the identification and enumeration of total CD19+ (B cells) and CD3-/CD56+ (NK cells) lymphocyte percentages and absolute counts in whole blood flow cytometry. The total lymphocyte percentage can obtained as well. | |
| Procode | GKZ | same | same |
| Regulation | 864.5220 | same | same |
| Sample input | MCL: 32 (12×72 mm) tubes
MPL: 24- or 96-well plates, or 12×72mm tubes | 32 (12×72 mm) tubes | 24- or 96-well plates, or 12×72mm tubes |
| Excitation | 488nm | same | same |
K182286
{5}
| Similarities | | | |
| --- | --- | --- | --- |
| Item | Device | Predicate (K030828) | Predicate (K071681) |
| Lasers | optional 633nm SSL or 635nm HeNe (not FDA-cleared) | | |
| Photodetector Circuits | FSC: APD
SSC: APD
FL1: PMT
FL2: PMT
FL3: PMT
FL4: PMT
FL5: PMT | same | same |
| Tarpon Amp Board Delay Line | potential signal loss, signal failure | same | same |
| Differences | | | |
| --- | --- | --- | --- |
| Item | Device | Predicate (K030828) | Predicate (K071681) |
| Detection of amp board delay line signal loss | automated pre-acquisition, post-acquisition | Manual inspection | Manual inspection |
| Disposition of compromised files | Quarantine of listmode data files suspected of compromise | Manual inspection | Manual inspection |
I. Special Control/Guidance Document Referenced (if applicable):
"Deciding When to Submit a 510(k) for a Software Change to an Existing Device", 25Oct17
"Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices", 11May05
Blue Book Memo K95-1: "510(k) Requirements During Firm-Initiated Recalls; Guidance on Recall and Premarket Notification Review Procedures During Firm-Initiated Recalls of Legally Marketed Devices", 21Nov95
J. Performance Characteristics:
1. Analytical Performance:
a. Accuracy:
To evaluate the effectiveness of the software detection tool in identifying cases of patient datafiles compromised by amp board signal defect, the Sponsor developed digital, electronic, and physical simulation methods to assess the accuracy of the software tool. For the digital simulation method, the LMD Tool edits listmode data
K182286
{6}
(LMD) files retroactively, simulating amp board signal loss through digital, algorithmic means. For the electronic simulation method, a modified amp board was engineered to include a switch on the delay line to electronically mimic the signal loss defect in compromised amp boards. For the physical simulation methods, the Sponsor physically disrupted the fluidic path in the instrument in an intermittent fashion. In addition, real-world data files were collected from instrument users who experienced amp board failures in their laboratory or generated from instruments that had known board failures.
For purposes of testing of the SDT, "positive" results are defined as files flagged by the SDT as compromised, and "negative" results are defined as files that were not flagged as compromised by the SDT. Therefore, "false positive" results are uncompromised files which the SDT flags, and "false negative" results are compromised files which are not flagged by the SDT.
i. Pre-Acquisition Testing:
To evaluate the ability of the software detection tool to detect pre-acquisition signal failures, the modified amp board was used to simulate signal loss conditions on each channel (FL1-5, FSC, SSC) using Tetra, Stem-Kit, and ClearLLab assays and auto-setup protocols. Out of a total of 247 tests across these parameters, 247 cases were positively identified by the SDT as compromised signal – the positive agreement rate for this value is 100.0% (95% CI: 98.5–100.0%).
ii. Post-Acquisition Test Case 1:
In the first set of test cases for post-acquisition performance of the SDT, the LMD Tool was applied to a set of 42 datafiles (Retrospective Simulations). In addition, the modified amp board or physically disrupted fluidics were employed (Live Simulations). Together, these methods generated 2344–2726 test datafiles for each assay panel. False negative (FNR) and false positive (FPR) rates were determined for the software detection tool. The FNR evaluates the SDT’s inability to identify a compromised file, and the FPR evaluates the frequency of falsely identifying an uncompromised file. The results are presented in the table below:
K182286
7
{7}
| | | | FNR1 | | | FPR2 | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Panel | Mode | Total n | n | point est. | 95% CI | n | point est. | 95% CI |
| Tetra 13 | FlowCount | 2344 | 15 | 0.65% | (0.40–1.08%) | 6 | 11.54% | (5.40–22.97%) |
| Tetra 2 | FlowCount | 2344 | 5 | 0.22% | (0.09–0.51%) | 5 | 9.62% | (4.18–20.61%) |
| Tetra 1 | relative | 2344 | 64 | 2.79% | (2.19–3.55%) | 0 | 0.00% | (0.00–6.88%) |
| Tetra 2 | relative | 2344 | 5 | 0.22% | (0.09–0.51%) | 1 | 1.92% | (0.34–10.12%) |
| StemKit | CXP | 2344 | 41 | 1.79% | (1.32–2.42%) | 4 | 7.69% | (3.03–18.17%) |
| B1 | ClearLLab | 2726 | 8 | 0.30% | (0.15–0.59%) | 1 | 1.92% | (0.34–10.12%) |
| B2 | ClearLLab | 2726 | 37 | 1.38% | (1.01–1.90%) | 1 | 1.92% | (0.34–10.12%) |
| M | ClearLLab | 2726 | 21 | 0.79% | (0.51–1.20%) | 2 | 3.85% | (1.06–12.98%) |
| T1 | ClearLLab | 2726 | 8 | 0.30% | (0.15–0.59%) | 2 | 3.85% | (1.06–12.98%) |
| T2 | ClearLLab | 2344 | 1 | 0.04% | (0.01–0.25%) | 3 | 5.77% | (1.98–15.64%) |
| 1 False Negative Rate: Compromised files not identified by software detection tool 2 False Positive Rate: Uncompromised file identified by software detection tool as compromised 3 Tetra 1 was considered to be representative of FlowCARE | | | | | | | | |
The results met the Sponsor's pre-determined acceptance criteria.
# iii. Post-Acquisition Test Case 2:
In the second set of test cases, challenging, difficult, or undetected compromised datafiles were modified by the LMD tool, and evaluated by the software detection tool. The ClearLLab assay was selected for this test case, as further demonstration of SDT performance at a greater degree of challenge.
| | | | FNR1 | | |
| --- | --- | --- | --- | --- | --- |
| Panel | Mode | Total n | n | point est. | 95% CI |
| B1 | ClearLLab | 31 | 0 | 0.0% | (0.0–11.0%) |
| B2 | ClearLLab | 29 | 0 | 0.0% | (0.0–11.7%) |
| M | ClearLLab | 36 | 3 | 8.3% | (2.9–19.0%) |
| T1 | ClearLLab | 39 | 1 | 2.6% | (0.5–13.2%) |
| T2 | ClearLLab | 22 | 0 | 0.0% | (0.0–14.9%) |
| 1 False Negative Rate: Compromised files not identified by software detection tool; no false positive evaluation (true negative) was performed | | | | | |
iv. Real World Data:
K182286
{8}
In order to evaluate the performance of the SDT in non-simulated conditions, sample listmode datafiles (LMD) were sourced from customers and end-users. A total of 409 datafiles were collected and categorized into three sets, as described below – by source and assay panel. These sets of datafiles were used to evaluate the performance of the SDT in detecting data compromised by actual amp board failure in laboratories using the affected instruments or prospectively by using a failed amp board in a single instrument.
| | | | | FNR1 | | | FPR2 | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Set | Source | Panel | Total n | FN/n | point est | (95% CI) | FP/n | point est. | (95% CI) |
| 1 | Customer complaints3 | various | 32 | 2/8 | 25.0%6 | (7.1-59.1%) | 1/24 | 4.2% | (0.7-20.2%) |
| 2 | Clinical site4 | StemKit | 233 | 0/8 | 0.0% | (0.0-32.4%) | 5/225 | 2.2% | (1.0-5.1%) |
| 3 | Re-engineered instrument5 | StemKit | 33 | 0/33 | 0.0% | (0.0-10.4%) | n/a | n/a | n/a |
| | | Tetra | 111 | 0/111 | 0.0% | (0.0-3.3%) | n/a | n/a | n/a |
| 1 False Negative Rate: Compromised files not identified by software detection tool
2 False Positive Rate: Uncompromised files identified by software detection tool as compromised
3 Listmode datafiles derived from customer complaint casefiles
4 Single clinical site with identified intermittent signal failure on one instrument
5 Compromised signal boards installed onto a single instrument for collection of 144 samples from 30 donors
6 FN from customer complaints were identified by manual review | | | | | | | | | |
The results met the Sponsor's pre-determined acceptance criteria.
v. Manual Review Reader Study:
The labeling addendum for use of the SDT specified that all data should be reviewed before release and reporting, irrespective of SDT flagging and quarantine. Therefore, instructions for manual review included in the labeling addendum constitute a critical mitigation for the SDT, and therefore a constituent element of the device.
To address the adequacy of these instructions for manual review, a reader study was performed. 100 data files from Test Case 1 were selected to represent all IVD assay areas (i.e. Tetra, StemKit, and ClearLLab), as well as all relevant types of signal loss and file compromise observed in the recall investigation, specifically signal loss patterns identified as resulting in false negative results from Test Cases 1 and 2. The datafiles included seven failure pattern-imprinted compromised LMD for each panel, and three unaltered "normal" files for each panel.
Analyses were provided to three Tier 1 independent reviewers, blinded to each other and to expected results. To model clinical practice, a second tier of review was further incorporated; to represent post-acquisition reviewers of higher level experience. Only discordant results among Tier 1 operators were resolved by Tier 2 review. Data for discordant results for a total of $n = 30$ reads for each panel are
K182286
{9}
presented in the table below.
| | | | | Discordance | | |
| --- | --- | --- | --- | --- | --- | --- |
| Panel | Mode | n | readers | n | point est. | (95% CI) |
| Tetra 1 | Flow-Count | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| Tetra 2 | Flow-Count | 10 | 3 | 1 | 3.3% | (0.6-16.7%) |
| Tetra 1 | relative | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| Tetra 2 | relative | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| StemKit | CXP | 10 | 3 | 1 | 3.3% | (0.6-16.7%) |
| B1 | ClearLLab | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| B2 | ClearLLab | 10 | 3 | 3 | 10.0% | (3.5-25.6%) |
| M | ClearLLab | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| T1 | ClearLLab | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
| T2 | ClearLLab | 10 | 3 | 0 | 0.0% | (0.0-11.4%) |
Discordant results were further reviewed by a second tier adjudicator, at a pathologist level of training. This secondary review corrected all but one of the discordant results (FN = 1/100: FNR 1.0% (0.2-5.5%).
b. Precision/Reproducibility:
Not applicable
c. Linearity:
Not applicable
d. Carryover:
Not applicable
e. Interfering Substances:
Not applicable
2. Other Supportive Instrument Performance Data Not Covered Above:
Not applicable
## K. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Parts 801 and 809, as applicable.
K182286
{10}
L. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
K182286
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.