K Number
K211305
Device Name
ANNE One
Manufacturer
Date Cleared
2021-09-14

(138 days)

Product Code
Regulation Number
870.2910
Panel
CV
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

ANNE One is a wireless vital signs and physiological data monitoring platform indicated for the measurement of heart rate, respiratory rate, step count, fall count, skin temperature, and body temperature by qualified healthcare professionals in healthcare settings. The device is intended for use on general care patients who are 18 years of age or older as a general patient monitor to provide continuous physiological information as an aid to diagnosis and treatment. The device is not intended for use on critical care patients. The device is not intended to monitor or measure respiratory rate or heart rate on ambulatory patients.

Device Description

ANNE One is a wireless multi-parameter vital signs monitoring system that consists of the following components, and subsystems:
A. ANNE Tablet (Samsung Galaxy Tablet) - manufactured by Samsung
B. ANNE View Mobile Application (software) manufactured by Sibel Inc.
C. ANNE Chest Sensor manufactured by Sibel Inc.
D. ANNE Limb Sensor manufactured by Sibel Inc.
E. ANNE Wireless Charger manufactured by Sibel Inc.
The ANNE Chest Sensor is a battery-operated, skin-mounted sensor with wireless transceiver capabilities worn on the upper body to measure heart rate, respiratory rate, step count, fall count, and skin temperature.
The ANNE Limb Sensor is an additional battery-operated skin-mounted sensor with wireless transceiver capabilities worn on the finger to measure skin temperature. When placed underneath the axillae, the ANNE Limb Sensor can measure body temperature.
Both the ANNE Chest and ANNE Limb Sensors continuously gather vital signs data from the person being monitored and then transmit the encrypted data to the ANNE Tablet operating the ANNE View Mobile Application-a mobile device software application.

AI/ML Overview

The provided text describes the ANNE One device, a wireless vital signs and physiological data monitoring platform, and its FDA 510(k) submission (K211305). While the document details various performance tests and validation methods, it does not provide specific acceptance criteria in a numerical format for each vital sign, nor does it present the reported device performance against such criteria in a table. It mostly states what was tested and how it was tested (e.g., according to specific ISO/IEC standards or against FDA-cleared reference devices).

Therefore, I cannot construct a table with acceptance criteria and reported device performance as requested, because that data is not explicitly present in the provided document. The document lists the types of measurements the device takes (heart rate, respiratory rate, step count, fall count, skin temperature, body temperature) and broadly states that bench and simulated use testing was conducted to demonstrate effectiveness and conformity to specifications.

However, I can extract information about the studies performed to demonstrate the device's capabilities:


Summary of Device Performance Studies for ANNE One (K211305)

The provided document describes various performance tests conducted to support the substantial equivalence determination for the ANNE One device. It does not explicitly state numerical "acceptance criteria" and "reported device performance" in a table format for each physiological parameter. Instead, it describes compliance with relevant standards and comparative studies against reference devices.

1. Table of Acceptance Criteria and Reported Device Performance

As mentioned, specific numerical acceptance criteria and reported performance values are not provided in the given text. The document broadly states that tests were performed to verify device specifications and ensure mechanical and electrical requirements were met, and that effectiveness was validated according to standards or against reference devices.

For example, for temperature measurement, it states: "Laboratory accuracy testing according to ISO 80601-2-56:2017 Section 201.101.2 validated the effectiveness of the ANNE Chest and Limb sensor for skin temperature measurements and the Limb Sensor for body temperature measurements when placed under the axillae." This indicates compliance with the standard but doesn't give a numerical accuracy. Similar statements are made for heart rate and respiratory rate.

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

  • Heart Rate:
    • Simulated Use Study: 35 healthy subjects.
    • Data Provenance: Not explicitly stated, but clinical studies are generally prospective. The phrase "healthy subjects" implies volunteers, likely from a single or a few sites.
  • Respiratory Rate:
    • Study: 50 healthy subjects.
    • Data Provenance: Not explicitly stated, but generally prospective for such studies.
  • Skin Temperature, Step Count, Fall Detection:
    • Simulated Use Testing: n=10 subjects.
    • Data Provenance: Not explicitly stated, but generally prospective.
  • Biocompatibility Testing: All patient-contacting materials were tested.
  • Electrical Safety & EMC Testing: Conducted on the device.
  • Software Verification & Validation Testing: Conducted on the software.
  • Animal Studies: Not required or performed.
  • Clinical Studies: Not required or performed for safety and effectiveness. (Instead, substantial equivalence was supported by Bench and Simulated Use testing).
  • Usability Testing: Conducted, but sample size not specified.

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

The document does not mention the use of experts to establish a "ground truth" in the context of human reader studies (e.g., for image interpretation). Instead, the ground truth for physiological measurements was established using FDA-cleared reference devices or manual observation.

4. Adjudication Method for the Test Set

Not applicable. The studies described are not "human-in-the-loop" studies requiring expert adjudication of reader performance. Measurements were compared against reference devices or manual counts.

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

No, an MRMC comparative effectiveness study was not performed, nor was it relevant to this device's function. The ANNE One is a vital signs monitoring device, not an imaging analysis AI intended to assist human readers in diagnosis. Therefore, there is no information on how human readers might improve with AI assistance. The performance studies focused on the accuracy of the device's measurements compared to established reference methods.

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

Yes, the performance studies described (accuracy of heart rate, respiratory rate, temperature, step count, fall count) essentially describe the standalone performance of the ANNE One system in measuring these vital signs. The device generates the data, and its accuracy is validated against gold standards (reference devices or manual counts). There isn't a human processing the algorithm's output for further interpretation in the way one might with an AI for medical image analysis.

7. The Type of Ground Truth Used

The ground truth for the vital signs measurements was primarily established using:

  • FDA-cleared reference devices: For heart rate (against an FDA-cleared reference device) and respiratory rate (against an FDA-cleared End Tidal Carbon Dioxide (EtCO2) reference device).
  • Reference standard/protocol conformance: For temperature (Laboratory accuracy testing according to ISO 80601-2-56:2017).
  • Manual counts: For step and fall counts, demonstrated to conform to design specifications through comparison to manual counts.

8. The Sample Size for the Training Set

The document does not specify any training set size. This is likely because the device is a hardware sensor system with associated software for data collection and display, rather than a machine learning model that requires a large separate training dataset in the typical sense of AI/ML-driven diagnostics. The "software verification and validation testing" refers to the traditional software development lifecycle testing, not machine learning model training.

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

As no training set is mentioned in the context of a machine learning model, this question is not applicable. The device's functionality is based on established physiological measurement principles and signal processing, not a trained AI model in the common sense.

§ 870.2910 Radiofrequency physiological signal transmitter and receiver.

(a)
Identification. A radiofrequency physiological signal transmitter and receiver is a device used to condition a physiological signal so that it can be transmitted via radiofrequency from one location to another, e.g., a central monitoring station. The received signal is reconditioned by the device into its original format so that it can be displayed.(b)
Classification. Class II (performance standards).