K070727 · Siemens Medical Solutions Diagnostics · JFY · Aug 3, 2007 · Clinical Chemistry
Device Facts
Record ID
K070727
Device Name
ADVIA CHEMISTRY ENZYMATIC CREATININE_2
Applicant
Siemens Medical Solutions Diagnostics
Product Code
JFY · Clinical Chemistry
Decision Date
Aug 3, 2007
Decision
SESE
Submission Type
Traditional
Regulation
21 CFR 862.1225
Device Class
Class 2
Indications for Use
The ADVIA Chemistry Enzymatic Creatinine_2 is for in vitro diagnostic use in the quantitative determination of creatinine in human serum, plasma, and urine on the ADVIA Chemistry Systems. Such measurements are used in the diagnosis and treatment of renal diseases, and in monitoring renal dialysis.
Device Story
ADVIA Chemistry Enzymatic Creatinine_2 is an in vitro diagnostic reagent system for quantitative creatinine measurement in human serum, plasma, or urine. Used on ADVIA Chemistry 1200/1650/1800/2400 systems; operated by laboratory personnel. Principle: enzymatic assay using creatininase, creatinase, sarcosine oxidase, and horseradish peroxidase. Sample mixed with two liquid reagents; creatinine converted to creatine, then hydrolyzed to sarcosine, then decomposed to form hydrogen peroxide. Peroxidase-catalyzed oxidative condensation with HMMPS and 4-aminoantipyrine produces blue pigment. Concentration determined via endpoint absorbance measurement. Output provides creatinine levels to clinicians for renal disease diagnosis, treatment, and dialysis monitoring.
Clinical Evidence
No clinical trials performed. Analytical performance established via bench testing. Precision evaluated per CLSI EP-5A2 (10-day study, 3 instruments). Linearity/reportable range (0.1–30 mg/dL serum; 1–245 mg/dL urine) validated using CAP linearity panels and NIST-traceable materials. Method comparison against predicate (n=44-60 per group) showed high correlation (r=0.99-1.00). Interference testing conducted for common endogenous substances and medications. IDMS reference method correlation (y = 0.981x - 0.01; r = 0.999) confirmed accuracy.
Indicated for quantitative determination of creatinine in human serum, plasma, and urine to aid in diagnosis and treatment of renal diseases and monitoring of renal dialysis. Prescription use only.
Regulatory Classification
Identification
A creatinine test system is a device intended to measure creatinine levels in plasma and urine. Creatinine measurements are used in the diagnosis and treatment of renal diseases, in monitoring renal dialysis, and as a calculation basis for measuring other urine analytes.
Predicate Devices
ADVIA® Chemistry Systems Creatinine_2 (k973993)
ADVIA® Chemistry Systems Enzymatic Creatinine (k991576)
Submission Summary (Full Text)
{0}
1
# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY TEMPLATE
A. 510(k) Number:
k070727
B. Purpose for Submission:
New Device
C. Measurand:
Creatinine
D. Type of Test:
Quantitative, enzymatic
E. Applicant:
Siemens Medical Solutions Diagnostics
F. Proprietary and Established Names:
Advia Chemistry Enzymatic Creatinine_2 (ECRE_2)
G. Regulatory Information:
1. Regulation section:
21 CFR § 862.1225 - Creatinine test system
2. Classification:
Class II
3. Product code:
JFY – Enzymatic Method Creatinine
4. Panel:
Clinical Chemistry (75)
H. Intended Use:
1. Intended use(s):
Refer to Indications for use below.
2. Indication(s) for use:
The ADVIA Chemistry Enzymatic Creatinine_2 is for in vitro diagnostic use in the quantitative determination of creatinine in human serum, plasma, and urine on the ADVIA Chemistry Systems. Such measurements are used in the diagnosis and treatment of renal diseases, and in monitoring renal dialysis.
3. Special conditions for use statement(s):
Prescription use only
4. Special instrument requirements:
For use with Siemens ADVIA 1200, ADVIA 1650 and ADVIA 2400 Chemistry Systems
I. Device Description:
The ADVIA Chemistry Enzymatic Creatinine_2 assay is available as a kit only. It consists of 2 reagents. Reagent 1 contains Creatinase (75 U/ml), Sarcosine oxidase (20 U/ml), and N-(3-sulfopropyl)-3-methoxy-5-methylaniline (0.9 mmol/L). Reagent
{1}
2 contains creatininase (400 U/ml), 4-aminoantipyrine (6.1 mmol/L), and peroxidase (50 U/ml).
# J. Substantial Equivalence Information:
1. Predicate device name(s): ADVIA Chemistry Enzymatic Creatinine (CREA_E) and ADVIA Chemistry Creatinine_2 (CREA_2)
2. Predicate $510(\mathrm{k})$ number(s): k991576 and k973993, respectively
3. Comparison with predicate:
| Item | New Device ADVIA Chemistry Enzymatic Creatinine_2 | Predicate - k991576 ADVIA Chemistry Enzymatic Creatinine |
| --- | --- | --- |
| Analyte | Creatinine | Creatinine |
| Method | Enzymatic (Creatininase) | Enzymatic (Creatinine Deiminase/GLDH) |
| Reagent components | Two liquid reagents: Reagent 1: Creatinase (75 U/ml), Sarcosine oxidase (20 U/ml), and N-(3-sulfopropyl)-3-methoxy-5-methylaniline (0.9 mmol/L); Reagent 2: Creatininase (400 U/ml), 4-aminoantipyrine (6.1 mmol/L), peroxidase (50 U/ml). | Two lyophilized reagents requiring reconstitution: Reagent 1: TRIS Buffer (150 mmol/L); Reagent 1 mix: NADPH (8 mg); 2-oxoglutarate (56.5 mg); GLDH >2KU Reagent 2 mix: Creatinine deiminase (40 U) |
| Format | Liquid | Liquid |
| Calibration | Single point | Single point |
| Calibrator | Bayer Chemistry Calibrator (REF 09784096) | Bayer Chemistry Calibrator (REF 09784096) |
| Linearity/Assay range | Serum: 0.1 – 30.0 mg/dL Urine: 0.1 – 245 mg/dL | Serum: 0.0 - 30.0 mg/dL |
| Expected Values (Serum) | Males: 0.6 – 1.1 mg/dL Females: 0.5 – 0.8 mg/dL | Males: 0.9 – 1.3 mg/dL Females: 0.6 – 1.1 mg/dL |
| Expected Values (Urine) | Males: 800 - 2000 mg/day Females: 600 - 1800 mg/day | Not Applicable |
| Closed reagent stability | Until the expiration date when stored at 2-8°C | Until the expiration date when stored at 2-8°C |
| Open reagent (on-board) stability | 60 days | 21 days |
| Sample matrix | Plasma, serum, urine | Plasma, serum |
# K. Standard/Guidance Document Referenced (if applicable):
CLSI EP-5A2: Evaluation of Precision Performance of Clinical Chemistry Devices; Approved Guideline
{2}
CLSI EP17-A: Protocols for Determination of Limits of Detection and Limits of Quantitation; Approved Guideline
L. Test Principle:
Creatinine is converted to creatine by the action of creatininase. The creatine formed is hydrolyzed by creatinase to produce sarcosine, which is decomposed by sarcosine oxidase to form hydrogen peroxide. In the presence of peroxidase, the hydrogen peroxide formed yields a blue pigment by condensation with N-(3-sulfopropyl)-3-methoxy-5-methylaniline (HMMPS) and 4-aminoantipyrine. The creatinine concentration is obtained by measuring the absorbance of the blue color at 596/694 nm. The absorbance of the color is proportional to the creatinine concentration.
M. Performance Characteristics:
1. Analytical performance:
a. Precision/Reproducibility:
Precision was evaluated for both serum and urine following recommendations in CLSI EP-5A2. Studies were conducted at one site over 10-day period with each sample tested 2 times per run and 2 runs per day. Two levels of serum pools, 2 levels of serum-based controls, and 3 levels of urine control were tested. Precision was evaluated for 3 ADVIA instruments for which the results are given below.
ADVIA 1200
| | | Within-Run | | Total | |
| --- | --- | --- | --- | --- | --- |
| Specimen | Level (mg/dL) | SD CV (%) | | SD | CV (%) |
| Serum Pool | 1.29 | 0.01 | 0.6 | 0.01 | 1.0 |
| Control 1 | 1.73 | 0.01 | 0.8 | 0.02 | 1.1 |
| Serum Pool | 3.04 | 0.03 | 0.8 | 0.03 | 0.9 |
| Control 2 | 8.79 | 0.06 | 0.6 | 0.08 | 0.9 |
| Urine | 42.34 | 0.30 | 0.7 | 0.41 | 1.0 |
| Urine | 79.78 | 0.78 | 1.0 | 0.91 | 1.1 |
| Urine | 133.09 | 1.22 | 0.9 | 1.34 | 1.0 |
{3}
ADVIA 1650
| Specimen | Level (mg/dL) | Within-Run | | Total | |
| --- | --- | --- | --- | --- | --- |
| | | SD | CV (%) | SD | CV (%) |
| Serum Pool | 1.29 | 0.01 | 0.8 | 0.01 | 1.1 |
| Control 1 | 1.76 | 0.01 | 0.5 | 0.02 | 0.9 |
| Serum Pool | 3.07 | 0.03 | 0.8 | 0.03 | 1.1 |
| Control 2 | 8.80 | 0.03 | 0.4 | 0.06 | 0.6 |
| Urine | 41.56 | 0.26 | 0.6 | 0.55 | 1.3 |
| Urine | 77.27 | 0.26 | 0.3 | 0.78 | 1.0 |
| Urine | 133.59 | 0.33 | 0.3 | 1.17 | 0.9 |
ADVIA 2400
| | | Within-Run | | Total | |
| --- | --- | --- | --- | --- | --- |
| Specimen | Level (mg/dL) | SD | CV (%) | SD | CV (%) |
| Serum Pool | 1.28 | 0.01 | 1.0 | 0.01 | 1.1 |
| Control 1 | 1.75 | 0.01 | 0.7 | 0.01 | 0.8 |
| Serum Pool | 3.06 | 0.03 | 0.8 | 0.03 | 0.9 |
| Control 2 | 8.81 | 0.10 | 1.1 | 0.12 | 1.4 |
| Urine | 41.13 | 0.15 | 0.4 | 0.41 | 1.0 |
| Urine | 77.54 | 0.18 | 0.2 | 0.69 | 0.9 |
| Urine | 131.33 | 0.40 | 0.3 | 1.23 | 0.9 |
b. Linearity/assay reportable range:
The sponsor reported linearity and reportable range data for ADVIA 1200, ADVIA 1650, ADVIA 1800 and ADVIA 2400 instruments. Two urine and serum sample pools (high and low) were used to prepare 9 equally spaced concentration levels for an approximate range of $3.5 - 30.0\mathrm{mg / dL}$ and $30 - 240\mathrm{mg / dL}$, respectively. Each test level was run in duplicate on the above analyzers. Based on a linear regression analysis conducted for measured and assigned values, Enzymatic Creatinine_2 assay is linear within the entire range tested across the entire group of instruments evaluated. The sponsor's acceptance criterion is $< 5\%$ difference between expected and observed values. The sponsor established the assay reportable range for urine at $1 - 245\mathrm{mg / dL}$. Since the above study for serum did not include the values for expected levels of creatinine, $(0.6 - 1.1\mathrm{mg / dL})$, the study below was conducted to evaluate the assay reportable range for serum.
To evaluate the linearity of the assay within the expected levels of creatinine in serum $(0.6 - 1.1\mathrm{mg / dL})$, the sponsor used the College of American
{4}
Pathologists (CAP) Creatinine Linearity/Accuracy Panel (LN24-A) designed to evaluate the accuracy, calibration, and linearity of a creatinine assays in the normal and elevated range. The data demonstrated that Enzymatic Creatinine_2 method on ADVIA 1200, ADVIA 1650, ADVIA 1800, and ADVIA 2400 instruments correlate with this panel and demonstrate the linearity of this method in the expected ranges. The data are presented in the table below. Based on the above studies, the limit of detection, and the method comparison study, the sponsor established the assay reportable range for serum and urine at $0.1 - 30\mathrm{mg / dL}$ and $1 - 245\mathrm{mg / dL}$ , respectively.
| CAP Creatinine LN24A Linearity/Accuracy Panel | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Sample | NIST Assigned Values | ADVIA 1200 ECRE_2 Recovery | % Bias | ADVIA 1650 ECRE_2 Recovery | % Bias | ADVIA 2400 ECRE_2 Recovery | % Bias |
| LN24-01 | 0.50 | 0.50 | 0.00 | 0.53 | 5.00 | 0.51 | 1.00 |
| LN24-02 | 0.74 | 0.71 | -4.05 | 0.73 | -2.03 | 0.72 | -3.38 |
| LN24-03 | 1.39 | 1.35 | -3.24 | 1.38 | -0.72 | 1.36 | -2.52 |
| LN24-04 | 2.05 | 1.97 | -3.90 | 2.03 | -1.22 | 2.00 | -2.68 |
| LN24-05 | 2.71 | 2.62 | -3.51 | 2.68 | -1.29 | 2.64 | -2.58 |
| LN24-06 | 3.36 | 3.28 | -2.38 | 3.34 | -0.74 | 3.27 | -2.83 |
| LN24-07 | 4.02 | 3.94 | -1.99 | 4.00 | -0.62 | 3.93 | -2.36 |
The sponsor also conducted extended linear range studies to establish system parameters that would trigger a re-run (reanalysis) of high serum or urine samples so that the labs can re-test samples without having to perform a manual dilution. The serum and urine samples were tested on all above mentioned ADVIA systems. Based on the sponsors' internal acceptance criteria of $\pm 10\%$ recovery (neat vs. diluted after reanalysis), the sponsor established the following upper limits of the extended ranges:
ADVIA 1650/1800/2400 serum/plasma: $5 \times 30.0 \, \mathrm{mg/dL} = 150 \, \mathrm{mg/dL}$
ADVIA 1650/1800/2400 urine: $5 \times 245 \, \mathrm{mg/dL} = 1225 \, \mathrm{mg/dL}$
ADVIA 1200 serum/plasma: $3 \times 30.0 \, \mathrm{mg/dL} = 90 \, \mathrm{mg/dL}$
ADVIA 1200 urine: $4 \times 245 \, \mathrm{mg/dL} = 980 \, \mathrm{mg/dL}$
c. Traceability, Stability, Expected values (controls, calibrators, or methods):
Calibrator. The sponsor uses single point calibration with previously cleared (k030169) Bayer Chemistry Calibrator supplied separately for both serum/plasma and urine. The assigned value of the calibrator is traceable to Isotope Dilution LC Mass Spectrometry reference method, which uses NIST reference material (SRM 967).
Controls. To ensure adequate quality control, the sponsor provides and recommends assaying two quality controls, along with the samples. The control materials have been previously cleared (k883209) and are supplied separately.
{5}
# d. Detection limit:
The sponsor evaluated Limit of Detection (LoD) and Limit of Blank (LoB) on all ADVIA instruments listed above following the guidelines in CLSI EP17-A. Based on the evaluation of 59 blank (saline) and 58 low-level (control 1 at 1.7 mg/dL) sample values, the LoD was established at $0.1\mathrm{mg / dL}$ following the algorithm, $\mathrm{LoD} =$ mean of blank $^+$ (1.6455 X SD of blank) $^+$ (1.645 X SD of Low sample).
# e. Analytical specificity:
The sponsor evaluated the effect of hemoglobin (0-1000 mg/dL), unconjugated bilirubin (0 - 30 mg/dL), conjugated bilirubin (0-30 mg/dL), and lipemia (intralipid) (0-1000 mg/dL) on normal serum control samples (1 mg/dL and 3 mg/dL) spiked with the interferents, and then compared with unspiked control. Based on the sponsor-defined interference limit of $\pm 10\%$ of control, the following interference limit claims were set by the sponsor for ADVIA 1200, ADVIA 1650/1800, and ADVIA 2400 instruments tested.
Interference results at $1\mathrm{mg / dL}$ creatinine
| Interferent | No Interference (±10%) claim up to (mg/dL) | | |
| --- | --- | --- | --- |
| | ADVIA 1200 | ADVIA 1650 | ADVIA 2400 |
| Hemoglobin | 500 | 750 | 500 |
| Lipemia | 1000 | 1000 | 1000 |
| Unconjugated bilirubin | 30 | 30 | 30 |
| Conjugated Bilirubin | 30 | 30 | 30 |
Interference results at $3\mathrm{mg / dL}$ creatinine
| Interferent | No Interference (±10%) claim up to (mg/dL) | | |
| --- | --- | --- | --- |
| | ADVIA 1200 | ADVIA 1650 | ADVIA 2400 |
| Hemoglobin | 1000 | 1000 | 1000 |
| Lipemia | 1000 | 1000 | 1000 |
| Unconjugated bilirubin | 30 | 30 | 22.5 |
| Conjugated Bilirubin | 30 | 30 | 30 |
Using an ADVIA 2400, the sponsor tested the following medications for interference at a creatinine level of $0.8 - 1.1\mathrm{mg / dL}$ .
| Medication | Medication Level (μg/mL) | Interference* |
| --- | --- | --- |
| Acetaminophen | 200 | NSI |
| Cefoxitin | 2230 | -11% |
| Cephalexin | 200 | NSI |
| Dipyrone | 200 | -13% |
| Dobesilate | 400 | -79% |
| Dobutamine | 20 | -10% |
| Dopamine | 20 | -15% |
| Ethylglycine | 6 | 13% |
| Fluorocytosine | 200 | NSI |
| L-dopa | 20 | NSI |
| Methyl dopa | 7 | NSI |
{6}
| Phenylbutazone | 200 | NSI |
| --- | --- | --- |
| Proline | 20 | NSI |
| Salicylate | 200 | NSI |
*NSI = No Significant Interference. A percentage effect ≥ 10% is considered a significant interference.
f. Assay cut-off:
Not Applicable
2. Comparison studies:
a. Method comparison with predicate device:
Performance of the ADVIA Chemistry Enzymatic Creatinine_2 assay was compared with the predicate device, ADVIA Chemistry Systems Creatinine_2 (CREA_2) (k973993). The ADVIA instrument systems 1200, 1650, and 2400 were used for the comparison. A summary of the sample number, composition, and the results of linear regression analysis is given below for serum and urine. The sponsor found correlation between the predicates and the new device for both serum and urine samples on ADVIA systems as follows:
| Specimen Type | Predicate/Instrument | n | Regression Equation | R | Sample range (mg/dL) |
| --- | --- | --- | --- | --- | --- |
| Serum | CREA_2/ADVIA1200 | 60 | y = 1.017x+0.03 | 1.00 | 0.3 – 12.3 |
| Serum | CREA_2/ADVIA1650 | 60 | y = 1.018x-0.04 | 1.00 | 0.3 – 11.9 |
| Serum | CREA_2/ADVIA2400 | 60 | y = 1.026x-0.03 | 1.00 | 0.3 – 12.1 |
| Urine | CREA_2/ADVIA1200 | 46 | y = 1.042x+0.41 | 1.00 | 20.0 – 238.1 |
| Urine | CREA_2/ADVIA1650 | 49 | y = 1.019x-0.99 | 1.00 | 17.8 – 239.4 |
| Urine | CREA_2/ADVIA2400 | 44 | y = 1.025x-2.47 | 0.99 | 18.9 – 218.1 |
In another method comparison study, serum samples were tested on all three instruments. The following results were obtained:
| Specimen Type | Predicate/Instrument | n | Regression Equation | R | Sample range (mg/dL) |
| --- | --- | --- | --- | --- | --- |
| Serum | CREA_2/ADVIA1200 | 28 | y = 1.047x+0.01 | 1.00 | 0.5 – 25.6 |
| Serum | CREA_2/ADVIA1650 | 42 | y = 1.059x-0.15 | 1.00 | 0.6 – 25.7 |
| Serum | CREA_2/ADVIA2400 | 42 | y = 1.054x-0.14 | 1.00 | 0.6 – 25.4 |
The sponsor also conducted a reference method correlation study using Isotope dilution mass spectrometry (IDMS), as recommended by the National Kidney Disease Education Program (NKDEP). Twenty five serum samples ranging in concentration from 0.5 – 4.5 mg/dL were assayed in three replicates within a day and found to produce the correlation equation (y = 0.981x -0.01; r = 0.999) between the device and the IDMS method.
b. Matrix comparison:
{7}
To demonstrate comparable performance between serum and lithium-heparin or potassium-EDTA plasma, the sponsor compared 42 samples on ADVIA instrument systems 1200, 1650, 1800 and 2400. Some samples were spiked with creatinine to extend the sample test range. Compared with the sera for the sample values ranged $(0.4\sim 26\mathrm{mg / dL})$ , the mean recovery for the lithium-heparin or potassium-EDTA plasma produced the following results:
| Specimen Type | Predicate/Instrument | n | Regression Equation | R | Sample range (mg/dL) |
| --- | --- | --- | --- | --- | --- |
| EDTA | ECRE_2/ADVIA1200 | 43 | y = 1.021x-0.03 | 0.998 | 0.4 – 28.2 |
| Heparin | ECRE_2/ADVIA1200 | 43 | y = 1.011x+0.04 | 0.999 | 0.4 – 28.6 |
| EDTA | ECRE_2/ADVIA1650 | 43 | y = 1.016x-0.02 | 0.999 | 0.3 – 25.6 |
| Heparin | ECRE_2/ADVIA1650 | 43 | y = 1.005x+0.05 | 0.999 | 0.4 – 26.0 |
| EDTA | ECRE_2/ADVIA2400 | 42 | y = 1.015x+0.0 | 0.998 | 0.4 – 26.8 |
| Heparin | ECRE_2/ADVIA2400 | 42 | y = 1.006x+0.07 | 0.999 | 0.4 – 27.3 |
# 3. Clinical studies:
a. Clinical Sensitivity: Not Applicable
b. Clinical specificity: Not Applicable
c. Other clinical supportive data (when a. and b. are not applicable):
# 4. Clinical cut-off: Not Applicable
# 5. Expected values/Reference range:
The expected values of creatinine were based on literature*. The sponsor states that they provide these ranges for reference only and each laboratory should establish its own normal range.
| Sex | Expected Values for Serum/Plasma | Expected Values for Urine |
| --- | --- | --- |
| Males | 0.6 – 1.1 mg/dL (53 – 97 μmol/L) | 800 – 2000 mg/day (7.1 – 17.7 mmol/day) |
| Females | 0.5 – 0.8 mg/dL (44 – 71 μmol/L) | 600 – 1800 mg/day (5.3- 15.9 mmol/day) |
* Tietz NW. Clinical Guide to Laboratory Tests. 4th ed. St. Louis, MO: WB Saunders Company; 2006:316.
# N. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.
# O. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.
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.