Search Filters

Search Results

Found 1 results

510(k) Data Aggregation

    K Number
    K232815
    Date Cleared
    2023-12-11

    (89 days)

    Product Code
    Regulation Number
    870.1130
    Reference & Predicate Devices
    Predicate For
    N/A
    AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
    Intended Use

    Wrist Blood Pressure Monitor is used to measure adult systolic blood pressure and pulse rate. The values are for diagnostic reference only. Suitable for medical institutions (such as hospitals, clinics, health centers, etc.) and home use.

    Device Description

    Wrist Blood Pressure Monitor mainly consist of pressure sensor, air pump, valve, cuff, main board and plastic case. Wrist Blood Pressure Monitor that uses the oscillometric principle to measure your blood pressure and pulse rate. The radial artery in the arm changes from blocked to open as the pressure in the cuff tied around the arm changes from high to low, causing the pressure in the cuff to be superimposed on a series of small pressure pulses. The sphygmomanometer senses these signals and, after certain calculations, finds the systolic and diastolic pressures of the radial artery in the body. The Wrist Blood Pressure Monitor include two models: YJ110, YJ111

    AI/ML Overview

    The provided text describes the non-clinical testing and a clinical validation study for a Wrist Blood Pressure Monitor (models YJ110, YJ111) to demonstrate its substantial equivalence to a predicate device (K223291).

    It is important to note that the document primarily focuses on non-clinical testing (e.g., appearance, safety indices, performance indices, electrical safety, EMC, software verification) and a clinical validation study that demonstrates the accuracy of the device against a reference method, rather than a comparative effectiveness study involving human readers with and without AI assistance. The device is a blood pressure monitor, not an AI-powered diagnostic tool requiring expert interpretation of images. Therefore, some of the requested information regarding AI-specific studies (e.g., MRMC studies, expert adjudication, AI standalone performance) would not be applicable in this context.

    Here's a breakdown of the available information:

    1. Table of Acceptance Criteria and Reported Device Performance

    The document provides a detailed table (Table VII-1) of "Inspecting item" with "Inspection content and acceptance standard" and the "Conclusion". Below is a summary, focusing on performance-related criteria:

    Inspecting Item / Acceptance StandardReported Device Performance (Conclusion)
    Appearance and Structure Test:Pass
    1.1 Correct shape, clean surface, no obvious scratches/damage/deformationPass
    1.2 Text/logo accurate, clear, firmPass
    1.3 Function clearly marked/indicatedPass
    1.4 Reliable connections, free key activityPass
    1.5 Display number clear, no missing/breakingPass
    Identification Requirement:Pass
    2.1 Device Identification (traceability, warnings, instructions, parameters)Pass
    2.2 External Packing (wrist circumference, battery info)Pass
    2.3 Specification (manual content, instructions, common arrhythmias, recalibration, measurement method, affected by posture, storage conditions, warranty, neonates, airbag risks, failure method, disinfection, correlation)Pass
    2.4 Low voltage prompt function (battery label, wristband ID)Pass
    Safety Index:Pass
    3.1 Maximum band pressure (<= 40kPa, pressure above 2kPa <= 3min)Pass
    3.2 Aerofluxus (deflation measure, time to drop 34.67kPa to 2kPa <= 10s)Pass
    Performance Index:Pass
    4.1 Range (Pressure: 0-39.3kPa, max 40kPa; Pulse: 40-199 times/min)Pass
    4.2 Resolution (Pressure: 0.133kPa; Pulse: 1 time/min)Pass
    4.3 Repeatability (difference <= 0.533kPa)Pass
    4.4 Pressure sensor and pulse measurement accuracy:Pass
    4.4.1 Pressure measurement max error: <= ±0.4kPa (±3mmHg)Pass
    4.4.2 Pulse measurement error: <= ±5%Pass
    4.5 Requirements for charging sources and pressure control valvesPass
    4.5.1 Aeration source (reach 40kPa in 200cm3 within 10s)Pass
    4.5.2 Pressure controlled air valvePass
    4.5.2.1 Air leakage (max pressure drop <= 0.133kPa in 10s)Pass
    4.5.2.2 Valve/wristband bleed rate (pressure reduction >= 0.267kPa/s)Pass
    4.5.2.3 Blow-by (time to drop 34.67kPa to 2kPa <= 10s)Pass
    4.6 Airbag and wristband requirementsPass
    4.6.1 Size (length 0.4x wrist, width 0.5x length)Pass
    4.6.2 Compression resistance (withstand 39.3kPa)Pass
    4.7 Wrist strap port/Structure (intact after 1000/10000 cycles)Pass
    4.8 System air leakage (rate of pressure drop <= 0.133kPa/s)Pass
    4.9 Lifetime (meet requirements after 10,000 full scale cycles)Pass
    4.10 Prompt facility (low battery, irregular heart rate warning)Pass
    4.11 Clock setting functionPass
    4.12 Memory function (60 sets)Pass
    4.13 Automatic shutdown function (after 3 minutes)Pass
    5. Electromagnetic compatibility (IEC60601-1-2)Pass
    6. Electrical safety (IEC 60601-1)Pass
    7. Environmental requirements test (IEC 60601-1-11)Pass

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

    The clinical validation study for the Wrist Blood Pressure Monitor (model YJ110, covering YJ111) used:

    • Sample Size: 86 adult subjects (47 females, 39 males).
    • Data Provenance: Not explicitly stated, but given the manufacturer is based in China, it is highly probable the study was conducted there.
    • Retrospective/Prospective: The text describes the study as "performed," indicating a prospective clinical validation study.

    3. Number of Experts Used to Establish Ground Truth and Qualifications

    This section is not applicable as the device is a blood pressure monitor, not an AI diagnostic tool requiring expert interpretation of medical images. The ground truth for blood pressure is established through a reference measurement method (e.g., auscultatory mercury sphygmomanometer in a supervised setting, following a recognized standard).

    4. Adjudication Method for the Test Set

    This section is not applicable for the same reason as above. Blood pressure measurement accuracy is typically assessed by comparing device readings to a reference standard, not through expert adjudication in the context of diagnostic interpretation. The study was performed using the "Same Arm Sequential Method" according to ISO 81060-2:2018.

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

    No, a MRMC comparative effectiveness study was not done. This type of study is relevant for AI-assisted diagnostic tools where human reader performance is being evaluated with and without AI assistance. This device is a direct measurement device (blood pressure monitor), not an AI-powered image interpretation device.

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

    Yes, the primary evaluation of the device's performance is essentially a standalone evaluation of the device's ability to accurately measure blood pressure. The clinical validation adhered to ISO 81060-2:2018, which is a standard for validating automated measurement type non-invasive sphygmomanometers. This standard assesses the device's measurement accuracy, which is its inherent algorithmic function for determining blood pressure and pulse rate from oscillometric signals.

    7. The Type of Ground Truth Used

    The ground truth for the clinical validation study was established by comparison to a reference measurement method per ISO 81060-2:2018. This standard specifies clinical procedures for validating the accuracy of automated non-invasive sphygmomanometers. Typically, this involves simultaneous or sequential measurements by the device under test and a calibrated reference sphygmomanometer (e.g., mercury sphygmomanometer) using an auscultatory method by trained observers. The "Same Arm Sequential Method" was chosen. The standard dictates the statistical analysis (mean error and standard deviation of differences) against the reference values.

    8. The Sample Size for the Training Set

    The document does not provide information on a specific "training set" sample size, as this device is not explicitly described as an AI/ML model that undergoes a distinct training phase. Its operation is based on the oscillometric principle with a non-invasive blood pressure (NIBP) algorithm, which implies a fixed, deterministic algorithm rather than a continuously learning or adaptable AI model. The "software function is justified by the software documents," as stated in "Different 2- Records Quantity" implies a validated software rather than a trained one.

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

    As there is no explicit mention of a training set or a distinct AI/ML training phase, this question is not directly applicable based on the provided text. The device uses an established "oscillometric principle" and "NIBP algorithm," suggesting a conventional engineering approach to device development and calibration, rather than a machine learning approach where "ground truth" is established for a training dataset. Accuracy is demonstrated via clinical validation against a reference standard.

    Ask a Question

    Ask a specific question about this device

    Page 1 of 1