LogoFDA 510(k) Search
Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K130919
    Device Name
    SENSUS
    Manufacturer
    NEUROMETRIX, INC.
    Date Cleared
    2013-07-02

    (90 days)

    Product Code
    GZJ
    Regulation Number
    882.5890
    Type
    Traditional
    Panel
    Neurology (NE)
    Reference & Predicate Devices
    K121059,K121305,K080950,K121184
    Why did this record match?
    Reference Devices :

    K121059, K121305, K080950

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

    The NeuroMetrix SENSUS is intended for use as a transcutaneous electric nerve stimulation device for the symptomatic relief and management of chronic intractable pain.

    The device may be used during sleep. The device is labeled for use only with the NeuroMetrix SENSUS Electrode.

    Device Description

    The SENSUS device, is a transcutaneous electrical nerve stimulator with a single output mode. The device utilizes a microprocessor running embedded software to control a high-voltage circuit that generates stimulating pulses with specific technical characteristics including pulse shape, amplitude (current), duration, pattern, and frequency. The device is powered by a permanent rechargeable Lithium-Ion battery that is charged through a USB cable connected to an AC adapter.

    The device delivers electrical stimulation to the patient through disposable, single-patient use electrodes placed on the patient's body. The device is labeled for use only with the SENSUS Electrode (K121816), to which it connects through insulated female medical snap connectors embedded within its housing.

    The device has a push-button that initiates stimulation, and controls the intensity. The device has a single two color LED for indication of stimulation status, battery charging, and error conditions.

    AI/ML Overview

    Here's an analysis of the acceptance criteria and study for the NeuroMetrix SENSUS device, based on the provided text:

    Acceptance Criteria and Device Performance

    The primary focus of the study described is the validation of the device's electrode peeling detection feature, which allows the device to be safely used during sleep.

    Acceptance Criteria CategorySpecific Acceptance CriteriaReported Device PerformanceOutcome
    Electrode Peeling DetectionFailure rate of 0%, with the upper bound of the one-sided 95% confidence interval for failure rate being below 5% (i.e., 0 out of 60 tests).0 failures out of 66 tests in two separate protocol runs (total 132 tests).Met
    Device must halt stimulation when remaining skin contact area is less than 3.5 cm².Average remaining contact area was 10.2 ± 2.1 cm² (range 6.9 - 19.5 cm²) for the 10-minute run and 10.1 ± 1.7 cm² (range 7.5 - 18.0 cm²) for the 40-minute run. All reported minimum remaining contact areas (6.9 cm² and 7.5 cm²) are greater than 3.5 cm².Met

    Study Details for Electrode Peeling Detection Validation

    1. Sample Size Used and Data Provenance:

      • Test Set Sample Size: 66 subjects and 132 SENSUS electrodes were used. There were two protocol runs, each with 66 tests, totaling 132 tests. The prospective validation sample size was set to 60 tests to meet the 0% failure rate statistical criterion.
      • Data Provenance: The study was described as "prospective validation," indicating it was specifically designed and conducted for this purpose. The country of origin is not explicitly stated, but NeuroMetrix is based in Waltham, MA, USA, suggesting the study was likely conducted in the USA.

    2. Number of Experts and Qualifications for Ground Truth:

      • The document does not explicitly mention the use of external experts to establish ground truth for the electrode peeling detection test. The "ground truth" for this test was the physical measurement of the electrode area remaining on the skin at the instant stimulation halted, as determined by the study methodology.

    3. Adjudication Method:

      • Not applicable/Not mentioned. The study involved objective measurements of physical parameters (electrode contact area).

    4. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:

      • No. This was a technical validation study for a safety feature, not a comparative effectiveness study involving human readers.

    5. Standalone (Algorithm Only) Performance:

      • Yes. The study focused on the automatic performance of the device's embedded algorithm in detecting electrode peeling and halting stimulation. There was no human-in-the-loop component for this safety feature.

    6. Type of Ground Truth Used:

      • Objective Measurement/Engineered Truth: The ground truth was established by direct measurement of the electrode-to-skin contact area at the point where the device halted stimulation. The "failure" condition was defined objectively as the remaining contact area being less than 3.5 cm².

    7. Sample Size for Training Set:

      • Not applicable/Not mentioned. The document describes a validation study for a specific safety mechanism (electrode peeling detection). It does not provide details of any machine learning model training or a "training set" in the context of AI. The device uses "embedded software to control a high-voltage circuit," implying rule-based or control system logic rather than a conventional machine learning model with a distinct training phase.

    8. How Ground Truth for Training Set was Established:

      • Not applicable, as no training set for a machine learning model is described. The device's operational parameters and safety thresholds (like the 3.5 cm² peeling threshold) would have been established through engineering design, risk analysis, and relevant standards, rather than a data-driven training process with a ground truth similar to clinical AI models.
    Ask a Question

    Ask a specific question about this device

    Page 1 of 1