Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K162382
    Date Cleared
    2017-04-14

    (233 days)

    Product Code
    Regulation Number
    862.1345
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Smart Dongle Blood Glucose Monitoring System consists of the Smart Dongle meter, single test strips, and the Smart Dongle mobile application as the display component of the Smart Dongle Blood Glucose Monitoring System. The Smart Dongle Blood Glucose Monitoring System is intended to be used for the quantitative measurement of glucose (sugar) in fresh capillary whole blood from the finger. This blood glucose monitoring system is intended to be used by a single person and should not be shared. Smart Dongle Blood Glucose Monitoring System is intended for selftesting outside the body (in vitro diagnostic use) by people with diabetes at home as an aid to monitor the effectiveness of diabetes control. This system should not be used for the diagnosis of or screening for diabetes, nor for use on neonates.

    Device Description

    The system consists of blood glucose meter, test strips and mobile platform (as a display of the system). And, the blood glucose meter is compatible to iPhone series, including iPhone 4, iPhone 4s, iPhone 5, iPhone 5s, iPhone 6, iPhone 6 plus, iPhone 6s plus. These products have been designed, tested, and proven to work together as a system to produce accurate blood glucose test results. Smart Dongle Blood Glucose Test Strips can be used only with the Smart Dongle Blood Glucose Monitoring System.

    AI/ML Overview

    This document is a 510(k) premarket notification for the Smart Dongle Blood Glucose Monitoring System. It describes the device, its intended use, and performance characteristics to establish substantial equivalence to a predicate device.

    Here's an analysis of the acceptance criteria and study information provided:

    1. Table of Acceptance Criteria and Reported Device Performance

    The acceptance criteria for the Smart Dongle Blood Glucose Monitoring System appear to be based on accuracy performance metrics, specifically comparing the device's readings to reference measurements. While explicit "acceptance criteria" are not phrased as such, the results presented imply the thresholds the device aims to meet.

    Acceptance Criteria (Implied)Reported Device Performance
    For glucose concentration < 75 mg/dL:
    Within 5 mg/dL42% (21/50)
    Within 10 mg/dL98% (49/50)
    Within 15 mg/dL100% (50/50)
    For glucose concentration ≥ 75 mg/dL:
    Within 5%51.8% (57/110)
    Within 10%95.5% (105/110)
    Within 15%100% (110/110)
    Precision (Intermediate Precision):
    Mean (mg/dL) for Level 1, SD (mg/dL), CV (%)49.6, 2.17, 4.38%
    Mean (mg/dL) for Level 2, SD (mg/dL), CV (%)139.4, 4.37, 3.13%
    Mean (mg/dL) for Level 3, SD (mg/dL), CV (%)335.4, 10.10, 3.01%
    Precision (Repeatability):
    Mean (mg/dL) for Level 1, SD (mg/dL), CV (%)49.7, 2.18, 4.39%
    Mean (mg/dL) for Level 2, SD (mg/dL), CV (%)91.2, 2.95, 3.23%
    Mean (mg/dL) for Level 3, SD (mg/dL), CV (%)128.3, 4.06, 3.16%
    Mean (mg/dL) for Level 4, SD (mg/dL), CV (%)227.8, 7.18, 3.15%
    Mean (mg/dL) for Level 5, SD (mg/dL), CV (%)390.2, 12.06, 3.09%

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

    • Sample size for accuracy testing:
      • For glucose concentration < 75 mg/dL: 50 measurements/samples.
      • For glucose concentration ≥ 75 mg/dL: 110 measurements/samples.
    • Data Provenance: Not explicitly stated. The document is from a Taiwanese company (TaiDoc Technology Corporation, New Taipei City, Taiwan). While the device is intended for the US market (FDA submission), the geographical origin of the clinical data (e.g., country where the samples were collected) is not specified. It is likely retrospective, as it's a 510(k) submission summarizing prior testing.

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

    Not applicable. For blood glucose monitoring systems, ground truth is typically established using a highly accurate laboratory reference method, not through expert human interpretation or consensus. The document does not mention human experts for ground truth.

    4. Adjudication Method for the Test Set

    Not applicable. Since the ground truth is established by a laboratory reference method for quantitative measurements, there is no need for human adjudication of results.

    5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done

    No. This is a medical device for quantitative measurement of blood glucose, not an imaging or diagnostic AI system requiring human reader performance studies.

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

    For a blood glucose monitoring system, the "standalone" performance is what is presented as the accuracy and precision data. The device itself (meter + test strip + mobile app for display) measures the glucose. The user's role is to apply the blood sample and read the result, not to interpret complex data that an algorithm might assist with. Therefore, the reported accuracy and precision data represent the standalone performance of the device without human interpretation of the measurement itself.

    7. The Type of Ground Truth Used

    The ground truth for blood glucose measurements is established by a laboratory reference method (e.g., a YSI analyzer or similar highly accurate enzymatic method). The document implicitly refers to this by providing comparison data without explicitly naming the reference method.

    8. The Sample Size for the Training Set

    The document does not provide information on a "training set" size. Blood glucose meters are typically developed and calibrated through extensive R&D and validation, but not in the same "training set" paradigm as machine learning algorithms where a specific dataset is used to train a model to learn patterns. The accuracy and precision studies presented are validation (test set) data for the final device.

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

    Not applicable for the reasons mentioned in point 8. If any internal calibration or development involved reference measurements, it would be against laboratory reference methods.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1