Search Results

Found 4 results

510(k) Data Aggregation

K Number

Validate with FDA (Live)

Device Name

NOVEOS Specific IgE (sIgE): Capture Reagent F013, Peanut (Arachis hypogaea); Capture Reagent F076, Bos d 4 a-lactalbumin, Milk; Capture Reagent F077, Bos d 5 ß-lactoglobulin, Milk; Capture Reagent F232, Gal d 2 Ovalbumin, Egg

Manufacturer

Hycor Biomedical

Date Cleared

2026-01-16

(136 days)

Product Code

Regulation Number

Type

Panel

Age Range

N/A

Reference & Predicate Devices

N/A

Predicate For

N/A

Intended Use

Device Description

AI/ML Overview

Ask a Question

Ask a specific question about this device

K Number

K252493

Validate with FDA (Live)

Device Name

NOVEOS Specific IgE (sIgE); Capture Reagent F024, Shrimp (Shrimp spp.); Capture Reagent F207, Clam (Mercenaria mercenaria); Capture Reagent F256, Walnut (Juglans spp.) ; Capture Reagent I006, German Cockroach (Blatella germanica)

Manufacturer

Hycor Biomedical

Date Cleared

2025-12-23

(137 days)

Product Code

Regulation Number

Type

Panel

Age Range

All

Reference & Predicate Devices

K200825,K051218

Predicate For

N/A

Intended Use

The NOVEOS Specific IgE Assay is an in vitro quantitative assay for the measurement of allergen specific IgE in human serum. NOVEOS Specific IgE Assay is to be used with the NOVEOS Immunoassay Analyzer. It is intended for use as an in vitro diagnostic aid in the clinical diagnosis of IgE mediated allergic disorders in conjunction with other clinical findings and is to be used in clinical laboratories.

Device Description

The NOVEOS Specific IgE Assay is an immunometric, chemiluminescent procedure for the quantitative determination of IgE of known specificity in human serum samples. It employs fluorescent labelled magnetic, streptavidin coated microparticles which are incubated with a biotinylated allergenic capture reagent, patient sample and monoclonal anti-human IgE antibody: horseradish peroxidase conjugate. If present in the sample, IgE binds to the biotinylated allergen captured to the streptavidin-coated microparticles to form a complex. After a final wash, the resulting complex is incubated with the enzyme substrate and a chemiluminescent signal is generated, the magnitude of which is proportional to the concentration of IgE in the patient sample.

The concentration of allergen-specific IgE is determined from a standard curve, which is traceable to the World Health Organization (WHO) reference reagent serum Immunoglobulin E (IgE) 11/234.

AI/ML Overview

N/A

Ask a Question

Ask a specific question about this device

K Number

K242531

Validate with FDA (Live)

Device Name

NOVEOS Specific IgE (sIgE) Assay; Capture Reagent G010, Johnson Grass (Sorghum halepense); Capture Reagent T007, Oak (Quercus alba) ; Capture Reagent G002, Bermuda Grass (Cynodon dactylon); Capture Reagent W001, Common Ragweed (Ambrosia artemisiifolia); Capture Reagent E005, Dog Dander (Canis familiaris); Capture Reagent T003, Common Silver Birch (Betula verrucosa); Capture Reagent F001, Egg White (Gallus gallus); Capture Reagent F002, Cow's Milk (Bos taurus); NOVEOS Immunoassay Analyz

Manufacturer

Hycor Biomedical

Date Cleared

2025-04-23

(240 days)

Product Code

Regulation Number

Type

Panel

Age Range

All

Reference & Predicate Devices

K200825,K182479,K051218

Predicate For

N/A

Intended Use

Device Description

The NOVEOS Specific IgE Assay is an immunometric, chemiluminescent procedure for the quantitative determination of IgE of known specificity in human serum samples. It employs fluorescent labelled magnetic, streptavidin coated microparticles which are incubated with a biotinylated allergenic capture reagent, patient sample and monoclonal anti‐human IgE antibody: horseradish peroxidase conjugate. If present in the sample, IgE binds to the biotinylated allergen captured to the streptavidin‐coated microparticles to form a complex. After a final wash, the resulting complex is incubated with the enzyme substrate and a chemiluminescent signal is generated, the magnitude of which is proportional to the concentration of IgE in the patient sample. The concentration of allergen‐specific IgE is determined from a standard curve, which is traceable to the World Health Organization (WHO) reference reagent serum Immunoglobulin E (IgE) 11/234.

AI/ML Overview

The provided FDA 510(k) clearance letter and summary for the NOVEOS Specific IgE (sIgE) Assay outlines the device's performance, but it does NOT describe "acceptance criteria" in an explicit, quantifiable manner that is typically found in a clinical study report. Instead, the document presents study results and compares them to a predicate device (ImmunoCAP Specific IgE) to demonstrate substantial equivalence, and also to clinical diagnosis of allergic status. The closest approximations to acceptance criteria are implicit in the performance metrics presented (e.g., target ranges for sensitivity, specificity, agreement, precision, linearity).

This device is an in vitro diagnostic (IVD) assay, not an AI/ML-based diagnostic imaging or analysis system. Therefore, the concepts of "human readers improve with AI vs without AI assistance," "standalone (algorithm only) performance," "number of experts," "adjudication method," and "training set ground truth establishment" do not directly apply in the same way they would for AI-powered diagnostic imaging devices. The "ground truth" for this IVD device is established through reference methods (Skin Prick Test, Oral Food Challenge, ImmunoCAP predicate device) or established clinical diagnosis.

Here's a breakdown of the information that is provided and how it relates to the requested points, with interpretations where necessary for IVD context:

1. Table of Acceptance Criteria and Reported Device Performance

As mentioned, explicit acceptance criteria are not stated. However, the performance data presented effectively serves as the "reported device performance" that presumably met the FDA's requirements for substantial equivalence. I will present the key performance metrics from the document.

Key Performance Metrics (Implicit Acceptance Criteria)

Performance Characteristic	Implicit Acceptance Criteria (based on predicate/clinical utility)	Reported Device Performance (NOVEOS sIgE Assay)
Clinical Sensitivity	Sufficient for clinical diagnostic aid (e.g., comparable to existing methods, supports clinical utility)	Varies by Allergen: - G010 (Johnson Grass): 72.9% (95% CI 62.7% to 81.2%)- T007 (Oak): 71.7% (95% CI 58.4% to 82.0%)- G002 (Bermuda Grass): 76.1% (95% CI 66.3% to 83.8%)- W001 (Common Ragweed): 62.0% (95% CI 50.3% to 72.4%)- E005 (Dog Dander): 71.9% (95% CI 54.6% to 84.4%)- T003 (Common Silver Birch): 55.1% (95% CI 41.3% to 68.1%)- F001 (Egg White): 52.8% (95% CI 37.0% to 68.0%)- F002 (Cow's Milk): 50.0% (95% CI 34.1% to 65.9%)Literature citation provided for lower sensitivity values to support observed performance.
Clinical Specificity	High, to minimize false positives	Varies by Allergen: - G010 (Johnson Grass): 99.2% (95% CI 95.9% to 99.9%)- T007 (Oak): 97.8% (95% CI 92.5% to 99.4%)- G002 (Bermuda Grass): 97.1% (95% CI 92.9% to 98.9%)- W001 (Common Ragweed): 93.6% (95% CI 86.8% to 97.0%)- E005 (Dog Dander): 100.0% (95% CI 95.2% to 100.0%)- T003 (Common Silver Birch): 100.0% (95% CI 93.2% to 100.0%)- F001 (Egg White): 100.0% (95% CI 97.4% to 100.0%)- F002 (Cow's Milk): 100.0% (95% CI 97.2% to 100.0%)
Positive Agreement (vs. ImmunoCAP)	High, demonstrating comparability to predicate device	Varies by Allergen: - G010 (Johnson Grass): 84.7%- T007 (Oak): 83.8%- G002 (Bermuda Grass): 89.4%- W001 (Common Ragweed): 72.8%- E005 (Dog Dander): 91.7%- T003 (Common Silver Birch): 96.0%- F001 (Egg White): 89.6%- F002 (Cow's Milk): 64.5%
Negative Agreement (vs. ImmunoCAP)	High, demonstrating comparability to predicate device	Varies by Allergen: - G010 (Johnson Grass): 99.3%- T007 (Oak): 94.6%- G002 (Bermuda Grass): 96.8%- W001 (Common Ragweed): 95.0%- E005 (Dog Dander): 96.7%- T003 (Common Silver Birch): 96.1%- F001 (Egg White): 96.2%- F002 (Cow's Milk): 98.9%
Total Agreement (vs. ImmunoCAP)	High, demonstrating overall comparability	Varies by Allergen: - G010 (Johnson Grass): 92.5%- T007 (Oak): 88.5%- G002 (Bermuda Grass): 93.8%- W001 (Common Ragweed): 82.2%- E005 (Dog Dander): 94.1%- T003 (Common Silver Birch): 96.0%- F001 (Egg White): 94.3%- F002 (Cow's Milk): 86.9%
Precision (Total %CV)	Typically low %CV for quantitative assays (e.g., <15-20%)	Varies by Allergen and Sample Concentration: Generally <15% for most samples, some low-concentration samples up to 22.1% (G010, Sample 1) or 15.7% (W001, Sample 1). Overall acceptable.
Linearity (R² All Samples)	R² close to 1, indicating good linearity across the assay range	Varies by Allergen: - G010 (Johnson Grass): 0.997- T007 (Oak): 0.991- G002 (Bermuda Grass): 0.997- W001 (Common Ragweed): 0.998- E005 (Dog Dander): 0.990- T003 (Common Silver Birch): 0.995- F001 (Egg White): 0.995- F002 (Cow's Milk): 0.995
LoB, LoD, LoQ	Low values, indicating good analytical sensitivity	Varies by Allergen:- LoB: 0.02 - 0.04 kU/L- LoD: 0.04 - 0.08 kU/L- LoQ: 0.10 - 0.16 kU/L
Interference	No significant interference at specified concentrations	No significant interference found for specified endogenous and exogenous substances.
Cross-Reactivity	No significant interference from specific immunoglobulins	No significant interference found for IgA, IgD, IgG, and IgM.
Immunological Specificity (Competitive Inhibition)	Dose-dependent inhibition with specific allergen, no inhibition with unrelated	Greater than 50% inhibition with specific allergen, no inhibition with related/unrelated allergens.
Stability	Meets claimed shelf-life, on-board, and open-vial stability	Accelerated data supports 9-36 months shelf-life, 15 days on-board, 365 days open-vial for capture reagents (real-time ongoing).

2. Sample Sizes Used for the Test Set and Data Provenance

Clinical Performance Study (Clinical Sensitivity & Specificity):
- Sample Size: Ranges from 102 to 228 patients overall, depending on the specific allergen.
  - "Atopic" (positive) samples: 32 to 88 samples confirmed by skin-prick testing or oral food challenge.
  - "Non-Atopic" (negative) samples: 53 to 146 samples deemed negative by ImmunoCAP testing (result <0.35 kU/L).
- Data Provenance: Not explicitly stated (e.g., country of origin). The study is described as a "clinical study," which implies real-world patient samples. It's unclear if the study was retrospective or prospective, but the description of "confirmed by skin-prick testing or oral food challenge" suggests a direct clinical evaluation.
Method Comparison to ImmunoCAP:
- Sample Size: Ranges from 175 to 268 patient samples, depending on the specific allergen.
- Data Provenance: Not explicitly stated (e.g., country of origin). These are "patient samples," implying clinical origins. Retrospective or prospective nature is not specified.
Precision/Reproducibility:
- Sample Size: 80-86 replicates per sample for within-laboratory precision (4-5 samples per allergen).
- Sample Size: 50-240 replicates per sample for lot-to-lot imprecision (4 samples per allergen).
Linearity:
- Sample Size: Three sets of serially diluted human serum samples (low, mid, high concentration) for each allergen. Dilutions created up to 1/32.

3. Number of Experts and Qualifications for Ground Truth

For an IVD device like this, "experts" in the sense of radiologists interpreting images are not directly involved in setting the ground truth for the device's output. The ground truth is laboratory or clinical reference standards.

Clinical Study Ground Truth: "Allergic status (atopic) was confirmed by skin-prick testing or oral food challenge, and the other 53 to 146 samples were deemed negative (non‐atopic) by ImmunoCAP testing (result <0.35 kU/L)."
- Number of Experts: Not specified. Skin-prick testing and oral food challenges are clinical procedures typically performed and interpreted by allergists or clinicians skilled in allergy diagnosis. ImmunoCAP testing is a laboratory-based method.
- Qualifications of Experts: Implied to be medical professionals (allergists/clinicians) and laboratory personnel capable of performing and interpreting these established diagnostic tests. No specific "years of experience" or board certifications are mentioned.
Method Comparison Ground Truth: The ImmunoCAP Specific IgE Assay (K051218) is used as the comparative "ground truth." This is a legally marketed predicate device.

4. Adjudication Method for the Test Set

Adjudication Method: Not applicable for this type of IVD device. The ground truth is based on established clinical diagnostic procedures (skin-prick test, oral food challenge) and a comparative predicate laboratory assay (ImmunoCAP), not on multiple human interpretations requiring adjudication.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

MRMC Study: No, a multi-reader multi-case (MRMC) comparative effectiveness study was not done. This type of study is typically performed for diagnostic imaging devices where human readers interpret medical images, often with and without AI assistance. This device is an in vitro diagnostic assay that directly measures a biomarker in serum.

6. Standalone (Algorithm Only Without Human-in-the Loop) Performance

Standalone Performance: The reported performance of the NOVEOS sIgE Assay (clinical sensitivity/specificity, agreement with ImmunoCAP, precision, linearity, etc.) is the standalone performance of the assay system. It operates automatically to measure allergen-specific IgE levels. There isn't a "human-in-the-loop" aspect to its direct output generation; it provides a quantitative result. The interpretation of that result for clinical diagnosis happens after the device provides its measurement, "in conjunction with other clinical findings."

7. Type of Ground Truth Used

Clinical Study Ground Truth:
- Clinical Diagnosis: Allergic status was confirmed by skin-prick testing or oral food challenge (for atopic individuals). These are considered definitive clinical methods for allergy diagnosis.
- Predicate Device Result: Samples deemed non-atopic were confirmed by ImmunoCAP testing (result <0.35 kU/L).
Method Comparison Study Ground Truth:
- Predicate Device Result: The ImmunoCAP Specific IgE Assay results were used as the comparator.

8. Sample Size for the Training Set

Training Set: This device is a quantitative in vitro diagnostic immunoassay, not a machine learning or AI algorithm in the traditional sense that requires distinct "training" datasets for model development. The development of such assays involves extensive analytical validation (e.g., reagent optimization, calibration curve establishment, linearity, precision) using various characterized samples and controls, but this is different from an AI model's "training set." The document describes various validation studies, but does not refer to a "training set" for an algorithm.

9. How the Ground Truth for the Training Set Was Established

Ground Truth Establishment for Training Set: Not applicable as described above. The assay's performance characteristics (e.g., calibration, measurement range) are established through a rigorous and well-defined analytical validation process, typically using reference materials, characterized control samples, and patient samples with known concentrations/statuses, rather than through ground truth established for an AI "training set." The calibrators are traceable to the World Health Organization (WHO) reference reagent serum Immunoglobulin E (IgE) 11/234, which serves as a fundamental "truth" reference for IgE measurement.

Ask a Question

Ask a specific question about this device

K Number

K200825

Validate with FDA (Live)

Device Name

NOVEOS Specific IgE (sIgE), Capture Reagent Cat Dander - E001, Felis Domesticus; NOVEOS Specific IgE (sIgE), Capture Reagent Timothy Grass- G006, Phleum pratense

Manufacturer

Hycor Biomedical

Date Cleared

2020-06-26

(88 days)

Product Code

Regulation Number

Type

Panel

Age Range

All

Reference & Predicate Devices

k051218

Predicate For

N/A

Intended Use

Device Description

The NOVEOS Specific IgE Assay is an immunometric, chemilyminescent procedure for the quantitative determination of IgE of known specificity in human serum samples. It employs fluorescent labelled magnetic, streptavidin coated microparticles which are incubated with a biotinylated allergenic capture reagent, patient sample and monoclonal anti-human IgE antibody: horseradish peroxidase conjugate. After a final wash, the resulting complex is incubated with the enzyme substrate and a chemiluminescent signal is generated, the magnitude of which is proportional to the concentration of IgE in the patient sample. The concentration of allergen-specific IgE is determined from a standard curve, which is traceable to the World Health Organization (WHO) reference reagent serum Immunoglobulin E (IqE) 11/234.

AI/ML Overview

This document describes the validation of the NOVEOS Specific IgE (sIgE) assays for Cat Dander (E001, Felis domesticus) and Timothy Grass (G006, Phleum pratense). This is a quantitative in vitro diagnostic device, not an AI/ML powered device. Therefore, many of the typical acceptance criteria for AI models, such as MRMC studies, ground truth establishment by experts, and training set details, are not applicable.

The acceptance criteria for this device are based on its analytical and clinical performance when compared to a legally marketed predicate device (ImmunoCAP Specific IgE Assay) and clinical diagnosis.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are generally implied by the performance metrics reported and the statistical analysis acceptable for FDA clearance of an in vitro diagnostic device of this type. The key performance indicators for a quantitative assay would typically include linearity, precision (imprecision/reproducibility), agreement with a predicate device, and clinical concordance (sensitivity and specificity).

Performance Table for NOVEOS Specific IgE (sIgE) Assays

Performance Metric	Acceptance Criteria (Implied)	Reported Device Performance (E001 - Cat Dander)	Reported Device Performance (G006 - Timothy Grass)
Agreement with Predicate	To demonstrate substantial equivalence, percentages should be high (e.g., >90% for PPA and >95% for NPA, or similar ranges). Precision of CI is also important.	PPA: 91.8% (95% CI: 84.6% – 95.8%)NPA: 99.3% (95% CI: 96.1% to 99.9%)(Cut-off: 0.35 kU/L)	PPA: 86.4% (95% CI: 78.5% to 91.7%)NPA: 98.5% (95% CI: 94.6% to 99.6%)(Cut-off: 0.35 kU/L)
Clinical Performance	Clinical sensitivity and specificity should demonstrate adequate diagnostic accuracy. Typical acceptance ranges for IVDs might be >70-80% for sensitivity and >90% for specificity, depending on the clinical context.	Sensitivity: 75.7% (95% CI 64.5% to 84.2%)Specificity: 100% (95% CI 97.1% to 100%)	Sensitivity: 76.2% (95% CI 64.4% to 85.0%)Specificity: 99.2% (95% CI 95.6% to 99.9%)
Imprecision/Reproducibility	Total CV% should generally be low, typically <10% to 15% across relevant concentrations and study phases (within-run, between-run, between-day, lot-to-lot, site-to-site). Specific ranges depend on the analyte and assay.	Total CV% ranges from 7.0-10.5% (E001, within-lab imprecision)Lot-to-lot CV% ranges from 7.6-9.5%Site-to-site reproducibility CV% ranges from 5.2-13.9% (upper bounds for low concentration samples)	Total CV% ranges from 5.0-11.1% (G006, within-lab imprecision)Lot-to-lot CV% ranges from 5.2-11.2%Site-to-site reproducibility CV% ranges from 6.0-11.1% (upper bounds for low concentration samples)
Linearity	R2 value close to 1.0 (e.g., >0.99) and slope close to 1.0 with intercept close to 0, within the claimed measuring interval.	R2: 0.999Slope: 1.00 (95% CI: 0.98 to 1.01)Intercept: -0.37 (95% CI: -0.79 to -0.05)Range: 0.03 - 101.62 kU/L	R2: 0.997Slope: 1.02 (95% CI: 0.99 to 1.05)Intercept: 0.58 (95% CI: -0.36 to 1.51)Range: 0.06 - 104.99 kU/L
Detection Limit	LoQ (Limit of Quantitation) should be defined and acceptable for clinical use, typically 0.14 kU/L for specific IgE assays to distinguish very low positive from negative (this is often the action limit for clinical use). LoB and LoD values should be reported.	LoB: 0.02 kU/LLoD: 0.06 kU/LLoQ: 0.14 kU/L	LoB: 0.03 kU/LLoD: 0.06 kU/LLoQ: 0.14 kU/L
Interference/Cross-Reactivity	Interference from common endogenous/exogenous substances and cross-reactivity with other related/unrelated allergens or immunoglobulins should be demonstrated to be minimal (e.g., <=15% interference).	Less than or equal to 15% interference for listed substances.Non-detectable cross-reactivity with other human immunoglobulins.<=15% inhibition for related/unrelated allergens.	Less than or equal to 15% interference for listed substances.Non-detectable cross-reactivity with other human immunoglobulins.<=15% inhibition for related/unrelated allergens.
Reference Range (Expected Values)	The device should align with established clinical cut-off values (e.g., <0.35 kU/L for negative) and support the verification of expected values in a normal population.	132 healthy subjects tested <0.35 kU/L (all).	127 healthy subjects tested <0.35 kU/L (all).
Stability	Shelf-life and on-board stability should be demonstrated to support the claimed storage and use conditions.	Shelf-life: 18-48 months (accelerated data). Real-time ongoing, currently supports 16 months.On-board: 48 hours to 28 days depending on component.	Shelf-life: 18-48 months (accelerated data). Real-time ongoing, currently supports 16 months.On-board: 48 hours to 28 days depending on component.

2. Sample Sizes Used for the Test Set and Data Provenance

The document describes several test sets for different performance evaluations.

Agreement with Predicate Device (Comparative Study):
- E001: 242 clinical samples.
- G006: 238 clinical samples.
- Data Provenance: The document does not explicitly state the country of origin. It refers to "clinical samples" and "patient samples" without further geographical details. The studies are assumed to be retrospective as they involve collecting and testing existing clinical samples.
Clinical Performance Study:
- E001: 200 samples (70 samples with allergic status confirmed by skin-prick testing and clinical history, and 130 samples from healthy, non-atopic donors with no reported allergy).
- G006: 188 samples (63 samples with allergic status confirmed by skin-prick testing and clinical history, and 125 samples from healthy, non-atopic donors with no reported allergy).
- Data Provenance: Similar to the predicate comparison, the country of origin is not specified, and the studies appear to be retrospective.
Reference Range Verification:
- E001: 132 apparently healthy subjects.
- G006: 127 apparently healthy subjects.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts

For this type of in vitro diagnostic device, "experts" in the context of AI models (e.g., radiologists interpreting images) are not directly applicable.

For Agreement with Predicate: The ground truth is the result obtained from the predicate device (ImmunoCAP Specific IgE Assay), which is a previously FDA-cleared and legally marketed device.
For Clinical Performance: The ground truth for the clinical study was established by "skin-prick testing and clinical history". This implies that clinical professionals (e.g., allergists, physicians) made the diagnosis of "allergic status" or "non-atopic" based on standard medical procedures and patient history, rather than experts reviewing data specifically for the study.

The number and qualifications of these clinical professionals are not specified in the document, as it's a standard clinical diagnostic process rather than a specific panel of experts assembled for ground truth annotation.

4. Adjudication Method for the Test Set

Adjudication methods like 2+1 or 3+1 are typically used for subjective assessments (e.g., image interpretation by multiple human readers). For this quantitative in vitro diagnostic device:

For Agreement with Predicate: There is no specific "adjudication" in the sense of reconciling differing expert opinions. The comparison is directly between the quantitative results of the new device and the predicate device.
For Clinical Performance: The "allergic status" or "non-atopic" ground truth was established by clinical diagnosis (skin-prick testing and clinical history). While clinical diagnosis often involves physician judgment, the document does not describe a formal multi-reader adjudication process; it refers to the standard clinical standard of care.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

No, an MRMC comparative effectiveness study was not done. This is an in vitro diagnostic assay (a laboratory test) that measures a biomarker (specific IgE) in human serum. It is not an AI-powered image analysis device or a device that directly assists human readers in interpreting complex visual data. Therefore, the concept of human readers improving with AI assistance is not applicable.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done

Yes, the primary performance evaluation of this device is standalone. The NOVEOS Specific IgE assay is an automated in vitro diagnostic test performed on the NOVEOS Immunoassay Analyzer. Its performance (quantitative measurement of IgE) is assessed directly through analytical studies (linearity, imprecision, detection limits, interference, cross-reactivity) and comparison to a predicate device/clinical diagnosis, without human intervention in the measurement process itself. The result is a quantitative value used by clinicians in conjunction with other findings.

7. The Type of Ground Truth Used

For Agreement with Predicate: The ground truth was based on the results obtained from the legally marketed predicate device (ImmunoCAP Specific IgE Assay). This is a common approach for establishing substantial equivalence for new IVD assays.
For Clinical Performance: The ground truth for allergic status was established by clinical diagnosis, specifically "skin-prick testing and clinical history." This is considered an objective clinical standard for diagnosing allergic disorders.

8. The Sample Size for the Training Set

This document describes the validation of an immunoassay, not an AI/ML model that requires "training sets" in the conventional sense. The device is a chemical/biological assay system. Therefore, there is no "training set" in the context of machine learning.

The studies described are for validation/testing of the already developed assay.

9. How the Ground Truth for the Training Set Was Established

As stated above, there is no "training set" for an AI/ML model for this type of device. The assay's "performance characteristics" (e.g., reagent formulations, assay parameters) would have been optimized during the device development phase, which is an engineering/chemistry process, not an AI model training process that uses labeled ground truth data.

Ask a Question

Ask a specific question about this device

Page 1 of 1