Retrospective clinical samples were used to perform a method comparison study between the new device and the predicate device to establish substantial equivalence.
Bio-Rad VARIANT™ Sickle Cell Short Program (K924813)
Agreement in hemoglobin identification (F, A, S, D, C, E)
Indications for Use
The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended as a qualitative screen for the presence of hemoglobins F, A, S, D, C and E in eluates of neonatal blood collected on filter paper by high performance liquid chromatography (HPLC). The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended for Professional Use Only. For In Vitro Diagnostic Use. The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended for use only with the Bio-Rad VARIANT™ nbs Newborn Screening System.
Device Story
System uses HPLC to analyze neonatal dried blood spot samples; samples punched from filter paper into microwell plates; auto-sampler aspirates eluate; analytical cartridge with cation exchange resin separates hemoglobins via gradient elution (two buffers); dual-wavelength photometer (415 nm/690 nm) detects hemoglobin absorbance. GDM software processes data; generates reports with sample ID, retention times, peak heights/areas, and chromatograms. Optional pattern assignment feature applies literature-derived rules to identify hemoglobin profiles. Used in clinical laboratories by professional staff; output assists clinicians in screening for hemoglobinopathies. Benefits include automated, high-throughput qualitative screening for sickle cell and other hemoglobin variants in newborns.
Clinical Evidence
Bench testing only. Precision evaluated per NCCLS EP-05A/EP-05A2; within-run precision <1% for all hemoglobin peaks. Limit of detection for variants (S, D, C, E) confirmed at 1% peak area. Method comparison performed on 1025 retrospective neonatal dried blood spot samples; 99.8% agreement with predicate device. No clinical studies performed.
Indicated for qualitative screening of hemoglobins F, A, S, D, C, and E in neonates using dried blood spot samples collected on filter paper.
Regulatory Classification
Identification
An abnormal hemoglobin assay is a device consisting of the reagents, apparatus, instrumentation, and controls necessary to isolate and identify abnormal genetically determined hemoglobin types.
Special Controls
*Classification.* Class II (special controls). A control intended for use with an abnormal hemoglobin assay is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 864.9.
Predicate Devices
Bio-Rad VARIANT™ Sickle Cell Short Program (K924813)
Submission Summary (Full Text)
{0}
510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION
DECISION SUMMARY
DEVICE ONLY TEMPLATE
A. 510(k) Number: K051072
B. Purpose for Submission:
The Bio-Rad VARIANT™ nbs Sickle Cell Program system/device provides improvements over the a predicate chromatographic devices known as the: VARIANT™ Sickle Cell Short Program [K924813; cleared 01/14/93].
C. Analyte:
Hemoglobins F, A, S, D, C and E
D. Type of Test:
IVD Qualitative, Hemoglobin HPLC.
E. Applicant:
Bio-Rad Laboratories, Inc.,
Clinical Systems Division
4000 Alfred Nobel Drive
Hercules, CA, U.S.A, 94547-1803
F. Proprietary and Established Names:
Bio-Rad VARIANT™ nbs Sickle Cell Program
Hemoglobin variants determination by HPLC
G. Regulatory Information:
1. Regulation section:
21 CFR 864.7415 [Abnormal Hemoglobin Assay]
2. Classification:
Class II
3. Product Code:
GKA
4. Panel:
Hematology (81)
H. Intended Use:
1. Intended use(s):
The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended as a qualitative screen for the presence of hemoglobins F, A, S, D, C and E in eluates of neonatal blood collected on filter paper by high performance liquid chromatography (HPLC).
The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended for Professional Use Only. For In Vitro Diagnostic Use.
The Bio-Rad VARIANT™ nbs Sickle Cell Program is intended for use only with the Bio-Rad VARIANT™ nbs Newborn Screening System.
{1}
Page 2 of 10
2. Indication(s) for use:
This device, consisting of the reagents, apparatus, HPLC instrumentation, software and controls, is indicated for professional laboratory IVD use to isolate and identify genetically determined abnormal (S, D, C, E) and normal (F, A) hemoglobin types in neonatal blood samples.
3. Special condition for use statement(s):
For professional IVD use only in clinical laboratory.
4. Special instrument Requirements:
For Bio-Rad VARIANTnbs Newborn Screening System.
I. Device Description:
This complete device consists of the Bio-Rad VARIANT™ nbs Sickle Cell Program (VnbsSCP) reagent kit, the Bio-Rad VARIANT™ nbs Newborn Screening System (VNBSS) instrument, and the Bio-Rad Genetic Data Management Software (GDM). The VNBSS instrument consists of a VARIANTnbs Neonatal Auto Sampler (VNAS) module for microwell plates and a VARIANTnbs Neonatal Chromatography Station (VNCS) module containing the high performance liquid chromatography (HPLC) hardware. The VnbsSCP reagent kit includes a specific analytical HPLC cartridge containing cation exchange resin, as well as two (2) buffer reagents for establishing an HPLC gradient. The GDM software is designed to execute the VnbsSCP assay protocol on the VNBSS instrument using the VnbsSCP reagent kit components for the purposes of qualitatively screening for the presence of normal hemoglobins F and A, as well as the abnormal hemoglobins S, D, C and E from neonatal heel stick blood, as collected on filter paper that is punched and eluted with deionized water. The VNBSS processes each sample individually. An eluted sample is aspirated directly from a microwell plate in the VNAS module with the punched filter paper disc still present, and transferred into the sample loop in the VNCS module. The contents of the sample loop are subsequently injected into the flow path of the VNCS module. The hemoglobins of interest are retained on the analytical cartridge in the presence of Elution Buffer 1. The ionic strength is subsequently raised by adding increasing amounts of Elution Buffer 2. The pre-programmed gradient is designed to have the hemoglobins of interest elute from the cartridge with retention times that fall within pre-determined windows characteristic of known normal and abnormal hemoglobins. The eluted hemoglobins are sequentially detected with a dual-wavelength filter photometer, which monitors hemoglobin absorbance at 415 nm and corrects for any gradient induced absorbance changes at 690 nm. The software processed HPLC data is outputted in a printed report that contains: 1) sample identification, 2) date and time of analysis, 3) a table of peaks that includes: observed peak identification (hemoglobin type) name(s), retention time(s), peak height(s), peak area(s), and relative area percent(s), 4) total chromatogram area, 5) complete chromatographic display and 6) any error message(s) relating (if needed) to such data. Also reported is an optional "pattern assignment" for each hemoglobin based upon "pattern rules" derived from diagnostic hemoglobin literature.
{2}
Page 3 of 10
# J. Substantial Equivalence Information:
1. Predicate device name(s): Bio-Rad VARIANT™ Sickle Cell Short Program
2. Predicate K number(s): K924813 [cleared: 01/14/1993]
3. Comparison with predicate:
| Summary of Technological Characteristic - Similarities to Predicate Device | | |
| --- | --- | --- |
| Features | New Device: Bio-Rad VARIANT™nbs Sickle Cell Program | Predicate Device: Bio-Rad VARIANT™ Sickle Cell Short Program (K#924813) |
| Intended Use | The Bio-Rad VARIANTnbs Sickle Cell Program is intended as a qualitative screen for the presence of hemoglobins F, A, S, D, C and E in eluates of neonatal blood collected on filter paper by high performance liquid chromatography (HPLC). | The VARIANT Sickle Cell Short Program is designed as a qualitative screen for the presence of hemoglobins F, A, S, D, C and E in eluates of neonatal blood collected on filter paper by high performance liquid chromatography. |
| | For In Vitro Diagnostic Use. | For In Vitro Diagnostic Use. |
| | For Professional Use Only. | For Professional Use Only. |
| Target Population | Neonates. | Neonates. |
| Design – Assay principle | Cation exchange high performance liquid chromatography. | Cation exchange high performance liquid chromatography. |
| Design – Assay Detection | Heme absorbance at 415 nm with background correction at 650 nm. | Heme absorbance at 415 nm with background correction at 650 nm. |
| Design – Analytes Identified | Six retention time windows for hemoglobins F, A, E, D, S and C. | Six retention time windows for hemoglobins F, A, E, D, S and C. |
| Design – Sample Type | Neonatal dried blood spots on filter paper collection cards. | Neonatal dried blood spots on filter paper collection cards. |
| Design – Punched Disc | One 1/8” disc. | One 1/8” disc. |
| Design – Manual Worklists | Accepts manual worklists. | Accepts manual worklists. |
| Materials - Components | Elution Buffer 1. Elution Buffer 2. Wash Solution. Analytical Cartridge. Lyophilized Whole Blood Primer. Lyophilized Retention Time Marker 1 (FAES). Lyophilized Retention Time Marker 2 (FADC). | Elution Buffer 1. Elution Buffer 2. Wash Solution. Analytical Cartridge. Lyophilized Whole Blood Primer. Lyophilized Retention Time Marker 1 (FAES). Lyophilized Retention Time Marker 2 (FADC). |
| Performance – Precision | Peak retention time precision is <1% for all hemoglobin peaks. | Peak retention time precision is <1% for all hemoglobin peaks. |
| Compatibility with Environment | U.S. FCC EMI and E.U. EMC standard compliant. | U.S. FCC EMI and E.U. EMC standard compliant. |
| Human Factors | For in vitro diagnostic use. For professional use only. | For in vitro diagnostic use. For professional use only. |
| Energy Used | Auto-switching 110 V and 220 V. | 110 V and 220 V models. |
{3}
Page 4 of 10
| Summary of Technological Characteristic - Similarities to Predicate Device | | |
| --- | --- | --- |
| Features | New Device: Bio-Rad VARIANT™nbs Sickle Cell Program | Predicate Device: Bio-Rad VARIANT™ Sickle Cell Short Program (K#924813) |
| Chemical Safety | Sodium azide concentration <0.05%. Gentamicin Sulfate concentration <0.1%. Tobramycin concentration <0.1%. Warnings provided in labeling as required, including State of California Proposition 65 Warning. | Sodium azide concentration <0.05%. Gentamicin Sulfate concentration <0.1%. Tobramycin concentration <0.1%. Warnings provided in labeling as required, including State of California Proposition 65 Warning. |
| Electrical, Mechanical and Thermal Safety | System certification to US and Canadian product safety standards and EU low voltage safety standards. | System certification to US and Canadian product safety standards and EU low voltage safety standards. |
| Standards Met | • EN375:2002
• EN591:2001
• EN980:2003
• EN1658:1996
• EN13485:2003
• EN13640:2002
• EN14971:2001
• EN61010-1:2001
• EN61010-2-101:2002
• EN61326:2001 | • EN375:2002
• EN591:2001
• EN980:2003
• EN1658:1996
• EN13485:2003
• EN13640:2002
• EN14971:2001
• EN61010-1:2001
• EN61010-2-101:2002
• EN61326:2001 |
| Summary of Technological Characteristic - Differences to Predicate Device | | |
| --- | --- | --- |
| Features | New Device: Bio-Rad VARIANT™nbs Sickle Cell Program | Predicate Device: Bio-Rad VARIANT™ Sickle Cell Short Program (K#924813) |
| Design – System Configuration | Separate chromatography and auto sampler modules and separate PC workstation with software. | Single integrated unit with chromatography, auto sampler and software functionalities. |
| Design – Media | CD-ROM. | ROM Card. |
| Design – Container, Elution Volume, Reconstitution Volume, Sample Loop, Column loading | Plastic 96 microwell plate. 250 μL sample elution volume. 500 μL primer and retention time marker reconstitution volume. 10 μL sample loop. Column loading is 1/25 of eluted sample volume and 1/50 of reconstituted material. | Plastic sample vial. 500 μL sample elution volume. 1000 μL primer and retention time marker reconstitution volume. 20 μL sample loop. Column loading is 1/25 of eluted sample volume and 1/50 of reconstituted material. |
| Design – Aspiration Probe Tip, Punched Disc Disposition | Beveled aspiration probe tip. Dried blood spot punched disc left in microwell during sample aspiration. | Blunt aspiration probe tip. Dried blood spot punched disc removed before sample aspiration. |
| Design – Additional retention time windows. | Seven (7) additional retention time windows: F1, “Other (1)”, “Other (2)”, “Other (3)”, “Other (4)”, “Other (5)” and “Other (6)”. | Feature not available. |
| Design – Automated Worklists | Accepts automated worklists from spot punchers. | Feature not available. |
| Design – Pattern Assignment | Optional pattern assignment feature uses pattern rules derived from literature. | Feature not available. |
{4}
Page 5 of 10
| Summary of Technological Characteristic - Differences to Predicate Device | | |
| --- | --- | --- |
| Features | New Device: Bio-Rad VARIANT™nbs Sickle Cell Program | Predicate Device: Bio-Rad VARIANT™ Sickle Cell Short Program (K#924813) |
| Performance – Variants Limit of Detection | The limit of detection for S, D, C and E is 1% of the total area of the sample when the total area is 1.5 million microvolt·second. | Limit of detection for hemoglobins S, D, C and E is 1% of the total area when the total area of the sample is 1.0 million microvolt·second. |
| Performance – Guideline for Interpretation of Results | Total area must be between 900,000 to 6.3 million microvolt·second. | Total area should range from 1,000,000–3,000,000 microvolt·second. |
| Performance – Total Area Limit of Detection | 900,000 microvolt·second. | Not addressed in instruction manual. |
| Performance – Bilirubin interference | Bilirubin up to 20 mg/dL does not interfere. | Not addressed in instruction manual. |
| Performance – Triglyceride interference | Triglyceride up to 6000 mg/dL does not interfere. | Not addressed in instruction manual. |
| Performance - Eluate stability | Eluates are stable for 48 hrs on the cooled auto sampler and at 2-8 °C and stable for 24 hrs at 15-30 °C. | Eluates are stable for 24 hrs at 2-8 °C. |
| Guidances Met | FDA 2002 - General Principles of Software Validation; Final Guidance for Industry and FDA Staff.
FDA 1999 - Guidance for Off-the-Shelf Software Use in Medical Devices; Final.
FDA 1998 - Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices; Final.
FDA 2003 – Device Advice; Content of a 510(k) | Not addressed. |
| Standards Met | NCCLS EP-05A 1999.
NCCLS EP-05A2 2004. | Not addressed. |
K. Standard/Guidance Document Referenced (if applicable):
| Standards Met | EN375:2002 - Information supplied by the manufacturer with in vitro diagnostic reagents for professional use.
EN591:2001 – Instructions for use in vitro diagnostic instruments for professional use.
EN980:2003 - Graphical symbols for use in the labeling of medical devices.
EN1658:1996 - Requirements for marking of in vitro diagnostic instruments.
EN13485:2003 - Quality systems - Medical devices - Particular requirements for the application of EN ISO 9001:1994.
EN13640:2002 - Stability testing of in vitro diagnostic medical devices.
EN14971:2001 - Medical devices – Application of risk management to medical devices.
EN61010-1:2001 - Safety requirements for electrical equipment for measurement, control, and laboratory use - Part 1: General requirements
EN61010-2-101:2002 - Safety requirements for electrical equipment for measurement, control, and laboratory use - Part 2-101: Particular requirements for in vitro diagnostic (IVD) medical equipment.
EN61326:2001 - Electrical equipment for measurement, control and laboratory use - |
| --- | --- |
{5}
Page 6 of 10
| | EMC requirements • NCCLS EP05A 1999 – Evaluation of Precision Performance of Clinical Chemistry Devices. • NCCLS EP05A2 2004 – Evaluation of Precision Performance of Quantitative Measurement Methods. |
| --- | --- |
| Guidances Met | • FDA 2002 - General Principles of Software Validation; Final Guidance for Industry and FDA Staff. • FDA 1999 - Guidance for Off-the-Shelf Software Use in Medical Devices; Final. • FDA 1998 - Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices; Final. • FDA 2003 – Device Advice; Content of a 510(k) |
## L. Test Principle:
Cation exchange high performance liquid chromatography with visible light detection.
## M. Performance Characteristics (if/when applicable):
### 1. Analytical performance:
a. Precision/Reproducibility:
The Bio-Rad VARIANT™ Sickle Cell Program device was based on NCCLS Protocol EP-05A (Vol. 19, No. 2 [1999]) and EP-05A2 (Vol. 24, No. 25 [2004]). Two analytical runs were performed per day on 20 days for a total of 40 runs on each of 3 separate systems. Each run included 4 replicates of two retention time positional QC controls. Within-run precision and within-device precision (formerly total precision) were determined. The reported predicate device retention time precision protocol included the same two retention time positional QC controls. However, the protocol did not strictly conform to NCCLS Protocol EP5-A2 guidelines, nor was within laboratory (or within-device) precision reported, as this was strictly a qualitative assay. The results of within run precision for both predicate and new devices was less than 1% for Hemoglobins F, A, E, D, S and C, and are presented in the following tables:
| Average Retention Time Summary | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| | | RT Window | 0.59-0.71 | 0.79-0.89 | 0.95-1.03 | 1.03-1.13 | 1.15-1.25 | 1.63-1.77 |
| | | | Retention Time Within-Run Precision (CV %) | | | | | |
| Device | Replicates | Sample | Peak F | Peak A | Peak E | Peak D | Peak S | Peak C |
| New System 1 | 160 | QC Control 1 | 0.63 | 0.83 | 0.98 | | 1.21 | |
| New System 2 | 160 | QC Control 1 | 0.63 | 0.83 | 0.98 | | 1.21 | |
| New System 3 | 160 | QC Control 1 | 0.63 | 0.83 | 0.98 | | 1.20 | |
| New System 1 | 160 | QC Control 2 | 0.63 | 0.83 | | 1.08 | | 1.70 |
| New System 2 | 160 | QC Control 2 | 0.63 | 0.83 | | 1.08 | | 1.70 |
| New System 3 | 160 | QC Control 2 | 0.63 | 0.83 | | 1.08 | | 1.70 |
{6}
Page 7 of 10
| Retention Time Within Run Precision Summary | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Device | Replicates | Sample | Retention Time Within-Run Precision (CV%) | | | | | |
| | | | Peak F | Peak A | Peak E | Peak D | Peak S | Peak C |
| New System 1 | 160 | QC Control 1 | 0.3 | 0.3 | 0.4 | | 0.3 | |
| New System 2 | 160 | QC Control 1 | 0.5 | 0.4 | 0.3 | | 0.3 | |
| New System 3 | 160 | QC Control 1 | 0.4 | 0.4 | 0.3 | | 0.2 | |
| Predicate | 10 | QC Control 1 | 0.7 | 0.0 | 0.0 | | 0.0 | |
| New System 1 | 160 | QC Control 2 | 0.3 | 0.3 | | 0.2 | | 0.3 |
| New System 2 | 160 | QC Control 2 | 0.5 | 0.4 | | 0.3 | | 0.2 |
| New System 3 | 160 | QC Control 2 | 0.5 | 0.3 | | 0.3 | | 0.1 |
| Predicate | 10 | QC Control 2 | 0.6 | 0.0 | | 0.4 | | 0.3 |
| Retention Time Within Device Precision (Formerly Total Precision) | | | | | | | | |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Device | Replicates | Sample | Retention Time Within-Run Precision (CV %) | | | | | |
| | | | Peak F | Peak A | Peak E | Peak D | Peak S | Peak C |
| New System 1 | 160 | QC Control 1 | 0.3 | 0.4 | 0.5 | | 0.5 | |
| New System 2 | 160 | QC Control 1 | 0.6 | 0.6 | 0.4 | | 0.6 | |
| New System 3 | 160 | QC Control 1 | 0.6 | 0.6 | 0.6 | | 0.5 | |
| Predicate | Not Reported | | | | | | | |
| New System 1 | 160 | QC Control 2 | 0.3 | 0.4 | | 0.4 | | 0.2 |
| New System 2 | 160 | QC Control 2 | 0.7 | 0.6 | | 0.4 | | 0.3 |
| New System 3 | 160 | QC Control 2 | 0.6 | 0.6 | | 0.5 | | 0.3 |
| Predicate | Not Reported | | | | | | | |
{7}
Page 8 of 10
b. Linearity/assay reportable range:
Refer to method correlation results in Section 2a,
c. Traceability (controls, calibrators, or method):
Not applicable
d. Detection limit:
The Bio-Rad VARIANT™ Sickle Cell Program new device peak area limit of detection for hemoglobin variants (E, D, S and C) was determined using a total of 102 sample measurements bracketing the 1% peak area limit of detection for each variant, when the total chromatogram area was 1.5 million microvolt x second (= μvolts·sec). The reported predicate device peak area limit of detection was 1%, when the total chromatogram area was 1.0 million μvolts·sec. The results re: peak area limit of detection for hemoglobin variants (E, D, S and C) was 1% for both predicate and new device, and is documented in the following table:
| Variant Peak Area – Limit of Detection | | | | | | |
| --- | --- | --- | --- | --- | --- | --- |
| Device | Chromatogram Total Area (microvolt-second) | Peak Area % Limit of Detection | | | | |
| | | Peak E | Peak D | Peak S | Peak C | |
| New System 1 | 1.5 | 1% | 1% | 1% | 1% | |
| Predicate | 1.0 | 1% | 1% | 1% | 1% | |
e. Analytical specificity:
Refer to method correlation results in Section 2a.
f. Assay cut-off:
Not applicable
2. Comparison studies:
a. Method comparison with predicate device:
Method correlation between predicate and -Rad VARIANT™nbs Sickle Cell Program new device was evaluated using 1025 unknown samples prepared from retrospective neonatal dried blood spot collection cards. Using a total area range of 0.90–6.30 million μvolts·sec, there was 99.8% (1023/1025) agreement between test and predicate devices for all sample hemoglobin identifications. There was 100.0% (250/250) agreement between test and predicate devices for identifying hemoglobin S (HbS) in those samples that predicate device identified as containing HbS. In the two disagreements, the new device identified all hemoglobins identified by the predicate as well as one additional peak in each case. [See explanation below.] These results indicate satisfactory correlation between the predicate and test devices. The results are presented in the following tables
{8}
Page 9 of 10
| Hemoglobin Identification Correlation Summary | | | | |
| --- | --- | --- | --- | --- |
| | Number of Samples | Predicate Device | New device | |
| | | F, A, E, D, S and (or) C Identified | Agree | Disagree |
| | 591 | FA | 590 | 1 (FAS) |
| | 52 | FAE | 52 | 0 |
| | 38 | FAD | 37 | 1 (FADC) |
| | 247 | FAS | 247 | 0 |
| | 90 | FAC | 90 | 0 |
| | 3 | FSC | 3 | 0 |
| | 2 | FC | 2 | 0 |
| | 1 | FE | 1 | 0 |
| | 1 | F | 1 | 0 |
| Total 1025 | | | 1023 | 2 |
In one case of disagreement the predicate device reported FA with an "unknown" peak in the S window while the test device reported FAS. Independent iso-electric focusing results identified FAS in support of the test device results.
| FA and FAS | Test Device | | |
| --- | --- | --- | --- |
| | | FA | FAS |
| Predicate Device | FA | 591 | 1 |
| | FAS | 0 | 247 |
In the other case of disagreement the predicate device reported FAD while the test device reported FADC with a $2.6\%$ peak area of HbC. Inspection of the test device chromatogram showed a low broad background in the C window without a resolved peak suggesting baseline noise.
| FAD and FADC | Test Device | | |
| --- | --- | --- | --- |
| | | FAD | FADC |
| Predicate Device | FAD | 37 | 1 |
| | FADC | 0 | 0 |
2. Comparison studies (continued):
b. Matrix comparison: Not applicable
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): Not applicable
4. Clinical cut-off: Not applicable
5. Expected values/Reference range: Not applicable
N. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence [SE] decision.
{9}
Page 10 of 10
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.