Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K222262

    Validate with FDA (Live)

    Device Name
    Sunrise
    Manufacturer
    Date Cleared
    2022-12-22

    (147 days)

    Product Code
    Regulation Number
    868.2376
    Age Range
    All
    Reference & Predicate Devices
    Predicate For
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticPediatricDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    The Sunrise device is a non-invasive home care aid in the evaluation of obstructive sleep apnea (OSA) in patients 18 years and older with suspicions of sleep breathing disorders.

    Device Description

    The Sunrise device is a cloud-based software device that analyzes data from a sensor (Sunrise sensor 1 or Sunrise sensor 2) placed on the patient's chin. The device detects respiratory events, identifies sleep stages and position. The device generates sleep parameters, e.g. apnea hypopnea index "Sunrise AHI", and position discrete states. Data collected by the device is integrated in a report for further interpretation by the healthcare provider.

    AI/ML Overview

    The provided text details the performance data for the Sunrise device to support its substantial equivalence determination. However, it does not explicitly state "acceptance criteria" in a表格 format as requested. Instead, it describes performance metrics (e.g., median measurement bias and LOA, sensitivity, specificity, global accuracy, and RMS values) for various parameters against pre-determined thresholds of clinical acceptability or against a gold standard (PSG).

    Based on the provided information, I will infer the acceptance criteria from the reported performance, as these are the values the device did achieve and were deemed sufficient for substantial equivalence.

    Here's a breakdown of the requested information:

    1. Table of Acceptance Criteria and Reported Device Performance

    As explicit acceptance criteria thresholds are not stated, the "Acceptance Criteria" column will reflect the reported performance that was deemed acceptable for substantial equivalence. The "Reported Device Performance" will reiterate these values.

    ParameterAcceptance Criteria (Inferred from Reported Performance)Reported Device Performance
    Study 1 (Belgium, n=289)
    TST Median Bias & LOAMedian bias within -4.50 min and LOA of -41.74 to +35.67-4.50 min (-41.74 to +35.67)
    AHI Median Bias & LOAMedian bias within -0.46 event/h and LOA of -13.52 to +9.00-0.46 event/h (-13.52 to +9.00)
    ORDI Median Bias & LOAMedian bias within +0.15 event/h and LOA of -10.70 to +10.12+0.15 event/h (-10.70 to +10.12)
    Sensitivity (AHI>=5)>= 0.990.99
    Sensitivity (AHI>=15)>= 0.920.92
    Sensitivity (AHI>=30)>= 0.810.81
    Specificity (AHI>=5)>= 0.860.86
    Specificity (AHI>=15)>= 0.940.94
    Specificity (AHI>=30)>= 0.990.99
    Study 2 (France, n=31)
    TST Median Bias & LOAMedian bias within -10.50 min and LOA of -37.42 to +25.79-10.50 min (-37.42 to +25.79)
    AHI Median Bias & LOAMedian bias within +0.20 event/h and LOA of -12.30 to +6.30+0.20 event/h (-12.30 to +6.30)
    ORDI Median Bias & LOAMedian bias within +1.01 event/h and LOA of -11.24 to +6.21+1.01 event/h (-11.24 to +6.21)
    Sensitivity (AHI>=5)>= 1.001.00
    Sensitivity (AHI>=15)>= 0.940.94
    Sensitivity (AHI>=30)>= 0.870.87
    Specificity (AHI>=5)>= 0.750.75
    Specificity (AHI>=15)>= 1.001.00
    Specificity (AHI>=30)>= 1.001.00
    Study 3 (Belgium, n=10)
    Position Discrete States Global Accuracy>= 93%93%
    Study 4 (SpO2 & Pulse Rate Accuracy)
    SpO2 Accuracy (RMS)<= 2.70% (over range 70-100%)2.70% (over range of 70-100%)
    Pulse Rate Accuracy (RMS)<= 1.95 bpm (over range 51-104 bpm)1.95 bpm (for a range of 51 to 104 bpm)
    Thermistor Ability to Capture AirflowPerformance equivalent to PSG oronasal thermal airflow sensorEquivalent to an oronasal thermal airflow sensor used in PSG

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

    • Clinical Study 1: 289 patients, retrospective, comparative, open study. Performed in Belgium.
    • Clinical Study 2: 31 patients, retrospective, comparative, open study. Performed in France.
    • Clinical Study 3: 10 patients, retrospective, comparative, open study. Performed in Belgium.
    • Clinical Study 4 (SpO2 & Pulse Rate): Not explicitly stated, but validated in accordance with ISO 80601-2-61:2019 and FDA guidance. This is typically a controlled bench study with a human subject population, but the document does not break down the sample size for this specific validation.
    • Thermistor Validation: Not explicitly stated (the text mentions "a validation study was conducted").

    All mentioned clinical studies are described as retrospective, comparative, and open studies.

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

    The document does not specify the number or qualifications of experts used to establish the ground truth (PSG data). It only refers to "the gold-standard PSG" as the comparison. In typical PSG studies, the PSG data is scored by trained sleep technologists and sometimes reviewed by a sleep physician, but this detail is not provided.

    4. Adjudication Method for the Test Set

    The document does not describe any specific adjudication method for the test set. The ground truth is stated to be "the gold-standard PSG."

    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 MRMC study comparing human readers with and without AI assistance is mentioned. The studies focus on the performance of the device (algorithm) itself against PSG.

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

    Yes, the clinical studies describe the performance of the "algorithm" and the "device" against PSG, indicating a standalone (algorithm only) evaluation. The text states: "The algorithm was used to analyze sensor data and evaluate the performance of the device compared to PSG."

    7. The Type of Ground Truth Used

    The primary ground truth used for OSA parameters (TST, AHI, ORDI, OSA severity) and position discrete states was Polysomnography (PSG), referred to as the "gold-standard PSG." For SpO2 and pulse rate accuracy, the ground truth was established in accordance with ISO 80601-2-61:2019 and relevant FDA guidance, which typically involves comparison against a reference oximeter or validated measurement system. For the thermistor, it was compared to an "oronasal thermal airflow sensor used in PSG."

    8. The Sample Size for the Training Set

    The document does not explicitly state the sample size for the training set. The clinical studies mentioned (n=289, n=31, n=10) are described as performance evaluation studies for the device, not necessarily for training. It states the "Sunrise algorithm... has been updated," implying a development process that would include training, but the specifics of the training dataset are not provided.

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

    The document does not provide information on how the ground truth for any potential training set was established. It focuses solely on the performance evaluation of the device against the "gold-standard PSG" for its validation studies.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1