K Number
K222879
Date Cleared
2023-01-24

(124 days)

Product Code
Regulation Number
882.5890
Panel
NE
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

The Transcutaneous Electrical Nerve Stimulator (model: TENS WMPS6-1) is an electrical nerve stimulator indicated for use for pain relief by applying an electrical current to electrodes on a patient's skin to treat pain. In particular, the TENS WMPS6-1 is indicated for use for symptomatic relief of chronic intractable pain and adjunctive treatment in the management of post-surgical and post traumatic pain.

Device Description

Transcutaneous Electrical Nerve Stimulator (model: TENS WMPS6-1) is an electrical nerve stimulator indicated for use for pain relief by applying an electrical current to electrodes on a patient's skin to treat pain. The device has 18 programs (13 standard programs and 5 editable programs), powered by 6 pieces of 1.5V batteries or AC 100-240V, comprising electronic stimulatory module and accessories of lead wires, electrodes. Six outlet sockets are used to connect skin electrodes by lead wires. The accessories of electrodes is 510(k) cleared device (K192568), Size: 50*50mm.

AI/ML Overview

The provided text is a 510(k) summary for a Transcutaneous Electrical Nerve Stimulator (TENS) device. It details the device's technical specifications and a comparison to a predicate device to establish substantial equivalence for regulatory clearance. This document does not describe an AI-powered medical device or an associated study for complex acceptance criteria like those involving AI performance metrics (e.g., sensitivity, specificity, MRMC studies, ground truth establishment by experts).

Therefore, I cannot extract the requested information regarding AI device acceptance criteria and study details from the provided text. The device is a physical medical device (TENS) and the "study" referred to in the document is a series of non-clinical bench tests to compare it to a predicate device, and to verify safety and performance standards.

Here's a breakdown of why I cannot fulfill your request based on the provided text:

  • No AI Component: The device is a Transcutaneous Electrical Nerve Stimulator, a physical product that applies electrical current. There is no mention of any AI or machine learning component.
  • No Diagnostic/Interpretive Task: The device is for pain relief, not for image analysis, diagnosis, risk prediction, or any other task that would typically involve acceptance criteria like sensitivity, specificity, or the need for ground truth established by experts.
  • No Human-in-the-Loop Study: Since there's no AI component, there's no MRMC (Multi-Reader Multi-Case) study described for human readers improving with AI assistance.
  • No Standalone Algorithm Performance: There is no algorithm to assess in a standalone capacity.
  • No Training Set/Test Set (in the AI context): The concepts of training and test sets as they apply to machine learning models are not relevant to the clearance of this physical device. The "test set" here refers to the device itself undergoing bench tests.
  • No Ground Truth Establishment by Experts: Ground truth for AI models is often established by expert consensus (e.g., radiologists reviewing images). For this TENS device, "ground truth" would refer to its physical and electrical properties meeting specified standards, validated through engineering tests, not human interpretation.

Summary of what the document DOES describe in terms of performance and comparison:

The document describes acceptance criteria in the context of substantial equivalence to an existing predicate device and compliance with voluntary standards for electrical medical equipment.

  • Acceptance Criteria (Implicit via Comparison to Predicate and Standards Compliance): The acceptance criteria are primarily demonstrated by showing that the new device's specifications (e.g., power source, output channels, waveform, voltage, current, safety features) are either identical, similar with acceptable differences (justified by safety and effectiveness analyses), or meet the requirements of relevant electrical safety and performance standards (ANSI AAMI ES60601-1, IEC 60601-1-2, IEC 60601-2-10). The reported device performance is presented as compliance with these standards and the specific electrical output values listed in the comparison table.

  • Study That "Proves" Compliance: The "study" here is a series of non-clinical bench tests.

    • Sample Size: This is not a study with a "sample size" in the statistical sense of patient data. It refers to testing the physical device itself.
    • Data Provenance: Not applicable as it's not a data-driven AI study. The tests are bench tests performed on the device.
    • Experts and Ground Truth: Not applicable in the way you've defined them for an AI study. Electrical engineers and regulatory compliance experts would be involved in setting and verifying the test standards.
    • Adjudication Method: Not applicable.
    • MRMC Comparative Effectiveness Study: No.
    • Standalone Performance: The "standalone" performance here refers to the device successfully passing all required electrical and safety tests independently (i.e., not reliant on human interpretation for its function).
    • Type of Ground Truth: The "ground truth" is compliance with established engineering and medical device safety standards for TENS devices (e.g., maximum current density limits, electrical isolation, waveform characteristics).
    • Training Set/Ground Truth for Training: Not applicable.

If you have a document describing an AI medical device, please provide that, and I will be able to answer your questions with the level of detail you are seeking.

§ 882.5890 Transcutaneous electrical nerve stimulator for pain relief.

(a)
Identification. A transcutaneous electrical nerve stimulator for pain relief is a device used to apply an electrical current to electrodes on a patient's skin to treat pain.(b)
Classification. Class II (performance standards).