cobas® Cdiff Nucleic acid test for use on the cobas® Liat® System (07454945190); cobas® Cdiff Positive and Negative Control Kit for use on the cobas® Liat® System (07454970190)
K260045 · Roche Molecular Systems, Inc. · OZN · Aug 28, 2026 · Microbiology
Device Facts
Record ID
K260045
Device Name
cobas® Cdiff Nucleic acid test for use on the cobas® Liat® System (07454945190); cobas® Cdiff Positive and Negative Control Kit for use on the cobas® Liat® System (07454970190)
Applicant
Roche Molecular Systems, Inc.
Product Code
OZN · Microbiology
Decision Date
Aug 28, 2026
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 866.3130
Device Class
Class 2
Indications for Use
The cobas Cdiff Nucleic acid test for use on the cobas Liat System is an automated, qualitative in vitro diagnostic test, that uses real-time polymerase chain reaction (PCR), for the detection of the toxin B (tcdB) gene of toxigenic Clostridioides difficile (C. difficile) in unformed (liquid or soft) stool specimens obtained from patients suspected of having C. difficile infection (CDI). The cobas Cdiff Nucleic acid test for use on the cobas Liat System is intended for use as an aid in the diagnosis of CDI in humans in conjunction with clinical and epidemiological risk factors.
Device Story
The cobas Cdiff Nucleic Acid test is an automated, qualitative in vitro diagnostic test for detecting the tcdB gene of toxigenic C. difficile. The system uses a single-use disposable assay tube containing all necessary reagents for sample purification, nucleic acid extraction, and real-time PCR amplification. The user adds a stool specimen to the tube, scans the barcode, and inserts it into the cobas Liat Analyzer. The analyzer automates the entire process, including reagent preparation, target enrichment, inhibitor removal, and real-time detection. The system uses the Roche Kinetic Algorithm v2 (KA2) to process fluorescence data and determine results. The device is intended for use in clinical settings as an aid in diagnosing CDI. It provides qualitative results to healthcare providers to assist in clinical decision-making and patient management.
Clinical Evidence
No new clinical trials were conducted. Evidence consists of re-analysis of raw fluorescence data from the original clinical performance study (K171770) using the updated KA2 algorithm. The re-analysis demonstrated consistency in clinical performance between the legacy and updated algorithms, confirming that primary acceptance criteria continue to be met.
Technological Characteristics
Real-time PCR assay; magnetic bead-based nucleic acid extraction; automated by cobas Liat Analyzer. Uses Roche Kinetic Algorithm v2 (KA2) and gLASP assay script (CDFA v1.4.0). Connectivity: Wired Ethernet and optional wireless via USB Wi-Fi adapter. Internal control: Bacillus thuringiensis israelensis. Single-use disposable assay tubes.
Indications for Use
Indicated for patients suspected of having Clostridioides difficile infection (CDI). Used for qualitative detection of toxin B (tcdB) gene in unformed (liquid or soft) stool specimens as an aid in diagnosis.
Regulatory Classification
Identification
A Clostridium difficile toxin gene amplification assay is a device that consists of reagents for the amplification and detection of target sequences in Clostridium difficile toxin genes in fecal specimens from patients suspected of having Clostridium difficile infection (CDI). The detection of clostridial toxin genes, in conjunction with other laboratory tests, aids in the clinical laboratory diagnosis of CDI caused by Clostridium difficile.
Special Controls
In combination with the general controls of the FD&C Act, the Portrait Toxigenic C. difficile Assay is subject to the following special controls: 1. The special controls for the Clostridium difficile toxin gene amplification assay are contained in the guidance document: "Class II Special Controls Guidance Document: Toxin Gene Amplification Assays for the Detection of Clostridium difficile."
*Classification.* Class II (special controls). The special controls are set forth in FDA's guideline document entitled: “Class II Special Controls Guideline: Toxin Gene Amplification Assays for the Detection of*Clostridium difficile;* Guideline for Industry and Food and Drug Administration Staff.” See § 866.1(e) for information on obtaining this document.
Predicate Devices
cobas Cdiff nucleic acid test for use on the cobas Liat System (K212427)
Submission Summary (Full Text)
{0}
[LOGO]
FDA
U.S. FOOD & DRUG
ADMINISTRATION
### 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION
DECISION MEMORANDUM
ASSAY AND INSTRUMENT
### I Background Information:
### A 510(k) Number
K260045
### B Applicant
Roche Molecular Systems, Inc.
### C Proprietary and Established Names
cobas Cdiff Nucleic acid test for use on the cobas Liat System (07454945190); cobas Cdiff Positive and Negative Control Kit for use on the cobas Liat System (07454970190)
### D Regulatory Information
| Product Code(s) | Classification | Regulation Section | Panel |
| --- | --- | --- | --- |
| OZN | Class II | 21 CFR 866.3130 – Clostridium Difficile Toxin Gene Amplification Assay | MI - Microbiology |
| OOI | Class II | 21 CFR 862.2570 – Instrumentation for clinical multiplex test systems | CH - Clinical Chemistry |
### II Submission/Device Overview:
### A Purpose for Submission:
To demonstrate that the Cobas Cdiff Nucleic acid Test run on a modified cobas Liat System with software version 4.0 is substantially equivalent to the previously cleared predicate device (i.e., K212427)
The updated cobas liat Analyzer Software (v4.0) incorporates the following changes:
- (a) Assay-related Changes: (i) update of cobas Cdiff assay-specific script from the legacy “CDFA v1.1.0” to “CDFA v1.4.0 generic Liat Assay Specific Packages (gLASP) assay script” and (ii) replacement of the legacy result calculation algorithm with a new Roche Kinetic Algorithm v2 (“KA2,” incorporated in the updated cobas Cdiff assay script).
- (b) Analyzer instrument-related Changes: (i) OS upgrade and (ii) enablement of optional wireless network connectivity.
### B Measurand:
Toxin B (tcdB) gene of toxigenic Clostridioides difficile (C. difficile)
Food and Drug Administration
10903 New Hampshire Avenue
Silver Spring, MD 20993-0002
www.fda.gov
{1}
# **C Type of Test:**
Real-time PCR assay for qualitative detection of *C. difficile* *tcdB* gene in unformed (liquid or soft) stool specimens
# **III Intended Use/Indications for Use:**
# **A Intended Use(s):**
See Indications for Use below.
# **B Indication(s) for Use:**
The cobas Cdiff Nucleic acid test for use on the cobas Liat System is an automated, qualitative *in vitro* diagnostic test, that uses real-time polymerase chain reaction (PCR), for the detection of the toxin B (*tcdB*) gene of toxigenic *Clostridioides difficile* (*C. difficile*) in unformed (liquid or soft) stool specimens obtained from patients suspected of having *C. difficile* Infection (CDI). The cobas Cdiff Nucleic acid test for use on the cobas Liat System is intended for use as an aid in the diagnosis of CDI in humans in conjunction with clinical and epidemiological risk factors.
# **C Special Conditions for Use Statement(s):**
Rx – For Prescription Use Only
# **D Special Instrument Requirements:**
cobas Liat Analyzer with cobas Liat Software v4.0
# **IV Device/System Characteristics:**
# **A Device Description:**
The cobas Cdiff Nucleic Acid test is a rapid, automated *in vitro* diagnostic (IVD) test for the qualitative detection and differentiation of *C. difficile* DNA in human stool specimens. The test is performed on the cobas Liat System which automates and integrates sample purification, nucleic acid amplification, and detection of the target sequence in biological samples using TaqMan probe-based real-time PCR assays. The system consists of an instrument and preloaded software for running tests and viewing the results. The system requires the use of a single-use disposable cobas Cdiff Nucleic Acid assay tube that holds all the sample purification and PCR reagents and hosts the sample preparation and PCR processes.
# **B Principle of Operation:**
The cobas Cdiff assay targets the oligonucleotide sequence of the *tcdB* (i.e., Toxin B gene of toxigenic *C. difficile*) that is implicated in *C. difficile* pathogenesis and allows differentiation between toxigenic and non-toxigenic *C. difficile* strains. The assay device includes a gram-positive bacterium (namely, *Bacillus thuringiensis israelensis*) which is co-processed with the target during the assay to extract nucleic acids and amplify oligonucleotide sequences for use as an Internal Control.
The test runs on the cobas Liat System, which consists of the cobas Liat Analyzer (hardware with integrated software) and single-use disposable assay tubes that contain all necessary sample preparation and PCR reagents in separate segments divided by frangible seals. The system fully automates all nucleic acid amplification test processes, including reagent preparation, target
K260045 - Page 2 of 8
{2}
enrichment, inhibitor removal, nucleic acid extraction, amplification, real-time detection, and result interpretation.
# C Instrument Description Information:
1. Instrument Name:
cobas Liat Analyzer with cobas Liat Software v4.0
2. Specimen Identification:
The cobas Liat system maintains traceability of specimen identification during processing and analysis by means of barcode labels present on each cobas Cdiff assay tube. Before starting the assay, the user is required to scan the assay tube barcode using the barcode reader integrated into the liat analyzer instrument. The specimen barcode is next scanned in, or a specimen identifier may be manually entered into the Liat system. The user adds the unformed (liquid or soft) stool specimen to the cobas Cdiff assay tube using a manufacturer-provided transfer pipette before closing the tube and re-scanning the cobas Cdiff assay tube barcode. The user next inserts the assay tube into the Liat analyzer and starts the run.
3. Specimen Sampling and Handling:
As indicated in the package insert, unformed (liquid or soft) stool specimens from patients with suspected C. difficile infection are collected in a clean, dry, and unused container by following institutional standard operating procedures. Prior to testing, each stool specimen is transferred to individual cobas PCR Media tubes using swabs provided in cobas PCR Media Uni Swab Sample kits. The cobas PCR Media tube inoculated with the swab is vortexed, and an aliquot of each stool specimen suspension is transferred to individual single-use disposable cobas Cdiff assay tubes using a manufacturer-provided transfer pipette and placed in the cobas Liat analyzer.
4. Calibration:
The Cobas Liat Analyzer hardware periodically performs automatic recalibration, and the Cobas Liat System does not run the assay until this auto-calibration step is performed.
5. Quality Control:
a. Internal Control: The assay tube contains a whole organism Internal Control (Bacillus thuringiensis israelensis ("Bti"), a chemically inactivated gram-positive bacterium), which is automatically added to all samples at the start of sample preparation for co-processing along with each sample. The cobas Liat Cdiff Internal Control checks for adequate processing of the target bacteria through all steps of the assay and monitors the presence of inhibitors in the sample preparation and PCR. The Internal Control is expected to be positive in a negative sample and can be negative or positive in a Cdiff positive sample. An invalid result will be generated if the Internal Control fails to meet the acceptance criteria.
b. Per lot Quality Control (QC) procedure: As indicated in the package, insert, the user is instructed to run a QC procedure (called the "Add Lot" procedure) before using a new lot of cobas Cdiff assay tubes. This procedure is required to validate the lot of cobas Cdiff assay tubes to be used and involves running a Cobas Liat Negative Control sample and a Cobas Liat Cdiff Positive Control sample. Valid results must be obtained for both Positive and Negative Control for the new lot of Cobas Cdiff assay tubes to be validated on the instrument.
K260045 - Page 3 of 8
{3}
c. Cobas Cdiff Positive Control: The assay Positive Control contains non-infectious DNA plasmids with C. difficile target sequence in buffer and is intended to verify the integrity of reagents in the Cobas Cdiff assay tube and proper function of the Cobas Liat Analyzer.
d. Cobas Liat Negative Control: The assay Negative Control contains no target and is intended to monitor potential target contamination in the workflow or environment.
# V Substantial Equivalence Information:
# A Predicate Device Name(s):
cobas Cdiff nucleic acid test for use on the cobas Liat System
# B Predicate 510(k) Number(s):
K212427
# C Comparison with Predicate(s):
| Candidate device & Predicate device: | Candidate: K260045 | Predicate: K212427 |
| --- | --- | --- |
| Device Trade Name | Cobas Cdiff Nucleic Acid test for use on the cobas Liat System | -same- |
| **General Device Characteristic Similarities** | | |
| Intended Use / Indications For Use | The cobas Cdiff Nucleic acid test for use on the cobas Liat System is an automated, qualitative *in vitro* diagnostic test that uses real-time polymerase chain reaction (PCR) for the detection of the toxin B (*tcdB*) gene of toxigenic *Clostridioides difficile* (*C. difficile*) in unformed (liquid or soft) stool specimens obtained from patients suspected of having *C. difficile* infection (CDI). The cobas Cdiff Nucleic acid test for use on the cobas Liat System is intended for use as an aid in the diagnosis of CDI in humans in conjunction with clinical and epidemiological risk factors. | -same- |
| Conditions for use | For Prescription Use only | -same- |
| Regulation Number | 21 CFR 866.3130; 21 CFR 862.2570 | -same- |
| Classification | Clostridium difficile Toxin Gene Amplification Assay Real-Time Nucleic Acid Amplification System | -same- |
| Product Code | OZN, OOI | -same- |
| Sample Type | Unformed soft stool specimens | -same- |
| Amplification Technology | Real-time PCR | -same- |
| Internal Control | A gram-positive *Bacillus thuringiensis israelensis* bacterial organism to monitor the full process of Cobas Liat Analyzer. Native sequence in the bacteria is used as the Internal Control Target. | -same- |
| Positive Control | Plasmid in buffer | -same- |
| Negative Control | Buffer only | -same- |
K260045 - Page 4 of 8
{4}
| Candidate device & Predicate device: | Candidate: K260045 | Predicate: K212427 |
| --- | --- | --- |
| Target Analyte | *C. difficile* Toxin B (*tcdB*) gene | -same- |
| Sample Collection Devices | Cobas PCR Media Swab Sample Kit | -same- |
| Sample Preparation | Magnetic bead-based nucleic acid extraction automated by the Cobas Liat Analyzer | -same- |
| Result Analysis | Based on PCR cycle threshold analysis | -same- |
| Assay Instrument | Cobas Liat Analyzer | -same- |
| **General Device Characteristic Differences** | | |
| System software and assay algorithm | ▪ Cobas Liat analyzer v4.0.0 ▪ Roche Kinetic Algorithm v2 (KA2) ▪ Generic Liat Assay Specific Packages (gLASP) assay script, CDFA v1.4.0 | ▪ Cobas Liat analyzer v3.3 ▪ Legacy algorithm ▪ Legacy IQuum Command Language (ICL) assay script, CFDA v1.1.0 |
| Wireless connectivity | ▪ Wired Ethernet connection ▪ Optional/alternative Wireless connectivity enabled via Off-The-Shelf USB Wi-Fi adapter | Wired Ethernet connection |
### VI Standards/Guidance Documents Referenced:
Class II Special Controls per 21 CFR §866.3130 Class II Special Controls Guideline Document: Toxin Gene Amplification Assays for the Detection of Clostridium difficile | FDA
### VII Performance Characteristics (if/when applicable):
#### A Analytical Performance:
The analytical performance of the Cdiff Nucleic Acid Test was evaluated following an update to the assay script (from the Legacy IQuum Command Language (ICL) assay script CDFA v1.1.0 to the generic Liat Assay Specific Package (gLASP) assay script CDFA v1.4.0). The updated script incorporates Roche's standardized Kinetic Algorithm 2 (KA2). To assess the impact of these changes, all analytical data previously submitted in support of the cleared cobas Liat Nucleic Acid Test device were reprocessed using the KA2 algorithm.
##### 1. Precision/Reproducibility:
In support of the candidate device with updated assay script and algorithm, raw fluorescence data from a within-laboratory precision study and a clinical reproducibility study (originally conducted in support of K171770 with legacy algorithm) were reprocessed using the updated algorithm. The re-analysis confirmed that the analytical precision remained consistent with the performance precision claims currently described in the package insert.
##### 2. Linearity:
Not applicable.
K260045 - Page 5 of 8
{5}
3. Analytical Specificity/Cross-reactivity, Interference, and Inclusivity:
Previously obtained raw fluorescence data from analytical specificity and inclusivity studies originally conducted for K171770 were reprocessed using the updated algorithm for the current candidate device. The reanalysis confirmed that the outcomes of cross reactivity, interference (endogenous and exogenous), and inclusivity (for forty (40) toxigenic C. difficile strains spanning multiple toxinotypes) studies remained consistent with the performance claims as described in the Package Insert.
4. Detection Limit:
Previously obtained raw fluorescence data from a limit of detection (LoD) study originally conducted for K171770 was reprocessed using the updated algorithm for the current candidate device (K260045). The reanalysis confirmed that analytical sensitivity/LoD remained consistent with the performance claims as described in the Package Insert.
5. Traceability, Stability, Expected Values (Controls, Calibrators, or Methods):
Metrological traceability is not applicable since there is no international standard for the Cdiff Nucleic Acid test.
The updated KA2 algorithm introduces a systematic shift in Ct values; therefore, manufacturing release specifications were recalibrated to maintain lot-to-lot performance consistency for the candidate device. Reanalysis of raw fluorescence data from all previously obtained real-time stability data for cobas Cdiff assay tubes was conducted with the updated algorithm, and no changes to the previously established 15-month shelf-life were observed.
6. Assay Cutoff:
The Cdiff Nucleic acid assay PCR Ct cutoffs and QC release specifications were revised for the transition to the updated KA2 algorithm (which differs in computational metrics from the legacy algorithm). To confirm that the algorithm update did not affect assay performance characteristics or cleared claims, all previously submitted data (originally processed under legacy CDFA v1.1.0) were reprocessed with CDFA v1.4.0, and no appreciable changes were observed in the final assay results.
# B Comparison Studies:
1. Method Comparison with Predicate Device:
To demonstrate performance equivalency of the Cdiff Nucleic acid assay following migration from the legacy assay script CDFA v1.1.0 running on software v3.5.1 to the updated assay script CDFA v1.4.0 running on software v4.0.0 and to verify all the changes to the cobas Liat system, a head-to-head equivalency study was conducted using ten (10) cobas Liat analyzer instruments, unique lots of positive and negative controls, as well as of swab kits and Cdiff assay tubes, and a contrived sample panel prepared with C. difficile ATCC 43255 at 2× and 5×LoD in cobas PCR media. In the study, the empaneled samples (along with media-only negatives) were tested on both the cleared and modified test systems. A total of 160 runs were performed (comprised of ten (10) replicates for each of negative and 5×LoD samples; forty (40) replicates at 2×LoD; and ten (10) replicates each of Positive and Negative controls samples, under each of the legacy and updated algorithms). All runs met the acceptance criteria and provided valid results, with no aborted or invalid runs. No deviations from the study protocol were observed.
K260045 - Page 6 of 8
{6}
The study concluded that migration from the legacy assay script to the updated gLASP assay script had no impact upon assay performance, confirming equivalency between the two script versions (as summarized in Table 1 below).
Table1: Equivalency study outcome for Cdiff target
| Test sample | CDFA Script v1.1.0 (legacy) | | | | CDFA Script v1.4.0 (gLASP) | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | Replicates | Target detected | Target not detected | Hit rate | Replicates | Target detected | Target not detected | Hit rate |
| Negative Control | 10 | 0 | 10 | 0% | 10 | 0 | 10 | 0% |
| Positive Control | 10 | 10 | 0 | 100% | 10 | 10 | 0 | 100% |
| Negative Sample | 10 | 0 | 10 | 0% | 10 | 0 | 10 | 0% |
| ~2× LOD (positive) Sample | 40 | 40 | 0 | 100% | 40 | 40 | 0 | 100% |
| ~5× LOD (positive) Sample | 10 | 10 | 0 | 100% | 10 | 10 | 0 | 100% |
#### Footnotes:
1. Out of 160 runs, all runs gave valid results. There were no aborted or invalid runs.
2. There were no deviations observed from the predetermined Equivalency Study Protocol.
3. All acceptance criteria outlined in Equivalency Study Protocol for cobas Liat assay's migration to gLASP Architecture were met.
#### 2. Matrix Comparison:
Not applicable.
### C Clinical Studies:
#### 1. Clinical Sensitivity:
Raw fluorescence data from a previous Clinical Performance /Clinical Utility Study (conducted to support the original clearance of the cobas Liat Cdiff Nucleic acid test, K171770) were reprocessed under the updated algorithm to support K260045. The reanalysis demonstrated consistency in clinical performance between the legacy and updated algorithms, with continued fulfillment of primary acceptance criteria and absence of any significant change to performance claims in the Package Insert.
#### 2. Clinical Specificity:
Please refer to section C1 above.
#### 3. Clinical Cut-Off
Not Applicable.
#### 4. Other Clinical Supportive Data (When 1. and 2. Are Not Applicable):
Not Applicable.
### D Expected Values/Reference Range:
Not Applicable.
### E Other Supportive Instrument Performance Characteristics Data:
EMC: Electrical safety and electromagnetic compatibility (EMC) testing were performed; the information was reviewed and found to be acceptable
K260045 - Page 7 of 8
{7}
**Software and Cybersecurity:** Software and Cybersecurity documentations were reviewed and found to be acceptable.
**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.
K260045 - Page 8 of 8
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.