Search Filters

Search Results

Found 2 results

510(k) Data Aggregation

    K Number
    K060688
    Date Cleared
    2006-04-13

    (29 days)

    Product Code
    Regulation Number
    864.7925
    Reference & Predicate Devices
    Why did this record match?
    Reference Devices :

    K951891, K002400, K033414

    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    HemosIL SynthASil is a high quality synthetic phospholipid reagent for the in vitro determination of Activated Partial Thromboplastin Time (APTT) in human citrated plasma on IL Coagulation and ELECTRA Systems.

    The product is used for the evaluation of the intrinsic coagulation pathway, APTT substitution test and the monitoring of heparin therapy.

    Device Description

    HemosIL SynthASil is a high quality synthetic phospholipid reagent for the in vitro determination of Activated Partial Thromboplastin Time (APTT) in human citrated plasma on IL Coagulation and ELECTRA Systems.

    The product is used for the evaluation of the intrinsic coagulation pathway, APTT substitution test and the monitoring of heparin therapy.

    AI/ML Overview

    This document describes the performance data for the HemosIL SynthASil device after optimization of its APTT parameter settings on the ACL Futura and ACL Advance systems. The stated purpose of this submission is to demonstrate substantial equivalence to the predicate device (K953981 HemosIL SynthASil) and to the HemosIL SynthASil reagent run on the ACL TOP system (K033414).

    Here's an analysis of the acceptance criteria and the study that proves the device meets them:

    1. Table of Acceptance Criteria and Reported Device Performance

    The document implicitly uses the performance of the predicate device (K953981 HemosIL SynthASil) and the HemosIL SynthASil on the ACL TOP (K033414) as the acceptance criteria for achieving "improved correlation" and "substantial equivalence." While explicit numerical acceptance criteria are not presented in a table format, the performance data provided aims to demonstrate this correlation.

    Performance MetricAcceptance Criteria (Implicit - from predicate/ACL TOP)Reported Device Performance (ACL Futura/Advance with optimized settings)
    Within Run PrecisionSimilar or equivalent precision to predicate device and/or HemosIL SynthASil on ACL TOP.Normal: Mean 28.5 seconds, CV% (Within run) 0.6, CV% (Total) 0.6
    Low Abnormal: Mean 49.0 seconds, CV% (Within run) 0.8, CV% (Total) 0.9
    High Abnormal: Mean 62.5 seconds, CV% (Within run) 0.7, CV% (Total) 0.8
    Method ComparisonStrong correlation with the predicate device/ACL TOP system, indicated by slope close to 1, intercept close to 0, and high correlation coefficient (r close to 1).Slope: 1.011, 0.984, 0.994, 0.973 (across 4 lots)
    Intercept: -0.664, 0.293, 0.159, 1.260 (across 4 lots)
    r: 0.9979, 0.9987, 0.9976, 0.9973 (across 4 lots)

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

    • Precision Test Set: The sample size for the precision study is not explicitly stated as a number of individual samples. It mentions "three levels of control plasma" and "assessed over multiple runs." This implies multiple measurements were taken for each of these three control levels.
    • Method Comparison Test Set: For each of the four SynthASil lots, n=93 or n=92 citrated plasma samples were used.
    • Data Provenance: The document does not specify the country of origin for the data. The study appears to be prospective in nature, performed to optimize and validate the device settings for regulatory submission. The samples are referred to as "human citrated plasma."

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

    This type of in-vitro diagnostic device (reagent for APTT determination) does not typically involve human expert interpretation for establishing ground truth in the same way an imaging AI device would. The "ground truth" for the method comparison is the measurement obtained from the predicate device or the ACL TOP system, which are themselves validated instruments. Therefore, there were no human experts establishing ground truth in this context.

    4. Adjudication Method for the Test Set

    Not applicable. As described above, this is an objective measurement study comparing an optimized device's performance to established methods, not subjective human interpretation requiring adjudication.

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

    • No MRMC study was done. This study concerns the performance of an in-vitro diagnostic reagent and instrument settings, not a device that involves human interpretation of results requiring a comparison of human reader performance with and without AI assistance.

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

    • Yes, this is effectively a standalone performance study. The document describes the performance of the HemosIL SynthASil reagent and its optimized settings on the ACL Futura and ACL Advance instruments. The results reported (precision, slope, intercept, r) are direct measurements from the device system without requiring human-in-the-loop analysis. The "algorithm" here refers to the optimized parameter settings and the reagent's performance characteristics.

    7. Type of Ground Truth Used

    • The ground truth used for the method comparison study was the measurements obtained from a legally marketed predicate device (K953981 HemosIL SynthASil) or an equivalent, legally marketed device (HemosIL SynthASil on the ACL TOP system - K033414). This is a form of "reference standard" or "comparative standard" established by existing, validated medical devices.

    8. Sample Size for the Training Set

    The document does not explicitly mention a "training set" in the context of machine learning or AI models. Instead, it refers to the "optimization" of APTT parameter settings. While this optimization process would have involved testing and refining these settings on various samples, the specific sample size, type, and method used for this optimization are not detailed in this summary. It's implied that the "optimization" process served a similar purpose to training, but no specific dataset is identified as a training set.

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

    Since there is no explicitly defined "training set" in the context of an AI/ML model, the concept of establishing ground truth for it is not directly applicable here. The "optimization" likely involved iterative adjustments of the parameter settings and testing them against known reference values or comparative measurements (similar to the method comparison's ground truth, but likely a larger or different set of samples used during development) to achieve the desired performance, particularly for correlation with the ACL TOP and predicate devices.

    Ask a Question

    Ask a specific question about this device

    K Number
    K043184
    Date Cleared
    2004-12-22

    (35 days)

    Product Code
    Regulation Number
    864.7750
    Reference & Predicate Devices
    Why did this record match?
    Reference Devices :

    K951891, K002400, K033414, K931721

    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    HemosIL RecombiPlasTin is a high sensitivity thromboplastin reagent based on recombinant human tissue factor (RTF) for the quantitative in virro diagnostic determination in human citrated plasma of:

    • Prothrombin Time (PT) on IL Coagulation and ELECTRA Systems .
    • Fibrinogen on IL Coagulation Systems only .
      The product is used for the evaluation of the extrinsic coagulation pathway and the monitoring of Oral Anticoagulant Therapy (OAT).
    Device Description

    HemosIL RecombiPlasTin is a high sensitivity thromboplastin reagent based on recombinant human tissue factor (RTF) for the quantitative in vitro diagnostic determination in human citrated plasma of:

    • Prothrombin Time (PT) on IL Coagulation and ELECTRA Systems .
    • . Fibrinogen on IL Coagulation Systems only
      The product is used for the evaluation of the extrinsic coagulation pathway and the monitoring of Oral Anticoagulant Therapy (OAT).
      The PT and fibrinogen parameter settings for HemosIL RecombiPlasTin on the ACL Futura and ACL Advance are being optimized for improved correlation with the ACL TOP, impacting the instrument-specific performance claims in the product insert.
    AI/ML Overview

    Here's an analysis of the provided text regarding the HemosIL RecombiPlasTin device, focusing on acceptance criteria, study details, and ground truth establishment:

    1. Acceptance Criteria and Reported Device Performance

    The acceptance criteria for this device appear to be primarily focused on its analytical performance metrics, specifically method comparison (correlation coefficient and slope) and precision (CV%). The document compares the new parameter settings on the ACL Futura/Advance to a legally marketed predicate device on the ACL TOP.

    MetricAcceptance Criteria (Implied by Predicate Performance)Reported Device Performance (HemosIL RecombiPlasTin with optimized parameters on ACL Advance)
    Prothrombin Time (PT)
    CV% (Within run)Not explicitly stated (but shown for predicate)Normal Control: 0.9%, Low Abnormal: 1.4%, High Abnormal: 1.4%
    CV% (Total)Not explicitly stated (but shown for predicate)Normal Control: 1.3%, Low Abnormal: 2.3%, High Abnormal: 3.9%
    Correlation Coefficient (r)Close to 1.0 (High correlation with predicate)0.9985
    SlopeClose to 1.0 (Agreement with predicate)1.039
    Fibrinogen
    CV% (Within run)Not explicitly stated (but shown for predicate)Not provided for Fibrinogen in this table
    CV% (Total)Not explicitly stated (but shown for predicate)Not provided for Fibrinogen in this table
    Correlation Coefficient (r)Close to 1.0 (High correlation with predicate)0.9811
    SlopeClose to 1.0 (Agreement with predicate)0.938

    Note: The exact numerical acceptance criteria are not explicitly stated as "must be greater than X" or "less than Y." Instead, the performance values of the legally marketed predicate device (HemosIL RecombiPlasTin on ACL TOP and HemosIL Fibrinogen-C on ACL TOP) serve as the benchmark for substantial equivalence. The reported values demonstrate very strong correlation and close agreement with the predicate for both PT and Fibrinogen, indicating the device meets the implied acceptance criteria for equivalence.

    2. Sample Size and Data Provenance for the Test Set

    • Sample Size for Test Set: 98 citrated plasma samples used for the method comparison study.
    • Data Provenance: Not explicitly stated (e.g., country of origin). The study involved "citrated plasma samples," which are human biological specimens. It's not specified if they were prospective or retrospective samples. As this is an in vitro diagnostic test for coagulation, the samples would likely be from a clinical setting, but further details are not provided.

    3. Number of Experts and Qualifications for Ground Truth of the Test Set

    This type of device (Prothrombin Time and Fibrinogen determination) relies on quantitative measurements, not subjective interpretation. Therefore, there are no "experts" in the traditional sense establishing a ground truth based on visual or interpretive assessment. The "ground truth" or reference method is the measurement obtained from the legally marketed predicate device (HemosIL RecombiPlasTin on ACL TOP for PT and HemosIL Fibrinogen-C on ACL TOP for fibrinogen).

    4. Adjudication Method for the Test Set

    Not applicable. As described above, the ground truth is established by a reference measurement from a predicate device, not through expert consensus or adjudication.

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

    Not applicable.

    • This is an in vitro diagnostic (IVD) device, specifically a reagent for automated coagulation analyzers.
    • It does not involve human readers interpreting images or data in a way that would necessitate an MRMC study.
    • The comparison is between two automated systems (the new device on ACL Futura/Advance vs. the predicate on ACL TOP), not between human readers with and without AI assistance.

    6. Standalone (Algorithm Only) Performance

    Yes, this is essentially a standalone performance study. The HemosIL RecombiPlasTin reagent, when run on the ACL Futura/Advance, performs the measurements automatically. The reported performance metrics (precision, slope, correlation) represent the analytical performance of the optimized reagent-instrument combination without human intervention in the measurement process itself. The "comparison" is the standalone performance of the new configuration against the standalone performance of the predicate configuration.

    7. Type of Ground Truth Used

    The ground truth used is a reference measurement from a legally marketed predicate device. Specifically:

    • For Prothrombin Time (PT), the reference was the HemosIL RecombiPlasTin on the ACL TOP.
    • For Fibrinogen, the reference was the HemosIL Fibrinogen-C on the ACL TOP.
      This is a common approach for demonstrating substantial equivalence for IVD devices, comparing the new device's measurements to those of an established, cleared device.

    8. Sample Size for the Training Set

    The document does not explicitly mention a separate "training set" in the context of machine learning or AI models. This device is a reagent with optimized parameters for existing instruments. The text states:
    "The PT and fibrinogen parameter settings for HemosIL RecombiPlasTin on the ACL Futura and ACL Advance are being optimized for improved correlation with the ACL TOP, impacting the instrument-specific performance claims in the product insert."

    This "optimization" process would involve internal development and testing to fine-tune the parameters. However, the specific size of the dataset used during this optimization phase is not provided in this 510(k) summary. The 98 plasma samples are explicitly described as being part of the method comparison study (which acts as the test set for regulatory submission).

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

    As there isn't a "training set" in the common AI/machine learning sense, the ground truth for any internal optimization would likely have been established in a similar manner to the test set: by running control plasmas and potentially patient samples on the predicate device (ACL TOP) to define the target reference values that the optimized parameters on the new instruments (ACL Futura/Advance) aimed to match. The goal was to achieve "improved correlation with the ACL TOP."

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1