K Number
K233755
Manufacturer
Date Cleared
2024-08-16

(268 days)

Product Code
Regulation Number
870.1425
Panel
CV
Reference & Predicate Devices
AI/MLSaMDIVD (In Vitro Diagnostic)TherapeuticDiagnosticis PCCP AuthorizedThirdpartyExpeditedreview
Intended Use

HeartKey Rhythm is intended for use by qualified healthcare professionals for analysing ECG data from adults. HeartKey Rhythm analyses a single ECG lead recorded from compatible devices used for arrhythmia diagnostics such as Holters, Extended Holters, ECG Patches, Event Recorders, or other similar devices when assessment of the rhythm is necessary.

The HeartKey Rhythm application program interface (API) is intended to be integrated into other device software to provide the following capabilities:

  • ECG signal conditioning,
  • ECG signal quality assessment,
  • QRS beat detection,
  • Ventricular ectopic beat detection,
  • ECG rhythm analysis

HeartKey Rhythm is not intended for use in life supporting, or sustaining systems, or real-time ECG monitors, or cardiac alarm devices, or OTC-use only devices. HeartKey Rhythm analysis results are not intended to be the sole means of diagnosis. It is offered to physicians on an advisory basis only in conjunction with the physician's knowledge of ECG patterns, patient background, clinical history, symptoms, and other diagnostic information.

Device Description

HeartKey Rhythm is intended to automatically analyse ECG data and provide ECG signal processing and assessment of cardiac arrhythmias. The software processes the following types of ambulatory ECGs:

  • Standard Holter: Lead I, II, III electrode configuration using gel electrodes e.g., Ag/AgCl,
  • Patch: Modified Lead II, Lead III, sternum using gel electrodes e.g., Ag/AgCl .

HeartKey Rhythm is capable of analysing both continuous ECGs as well as event recorders. The device provides the following analysis:

FeatureDescription
ECG Signal Processing Engine (SPE)Signal Conditioning (SC): Reduce the noise artefact present in the input ECG signal for physician review of NSR, AF and VE beats. Beat Detector (BD): Detect the QRS complexes of the input ECG signal. Signal Quality (SQ): Identify and classify the ECG segments' signal quality from the input ECG signal.
ECG Arrhythmia Analysis (AA)Rhythm Analysis: Normal Sinus Rhythm, Atrial Fibrillation, Cardiac Pause, Tachycardia, Bradycardia, Inconclusive, Unreadable. Beat Analysis: Normal (N), Ventricular (V) Ectopic. Beat Clustering: Beat Cluster Template, Beat Cluster Count, QRS Location, Beat Classification, Heart Rate

The Software Library may be integrated into ECG management systems, ECG analysis platforms or remote patient monitoring platforms. The filtered and enhanced ECG is not intended to be used by other automated analysis algorithms.

AI/ML Overview

The provided text describes the acceptance criteria and the study that proves the device meets those criteria for the HeartKey® Rhythm device.

Here's a breakdown of the requested information:

1. A table of acceptance criteria and the reported device performance:

The document states that "Performance testing included the evaluation of metrics such as sensitivity, specificity, and positive values. These outputs were evaluated against a clinically relevant acceptance criteria to demonstrate HeartKey Rhythm's effectiveness." However, the exact numerical values for the acceptance criteria and the reported device performance for these metrics (sensitivity, specificity, positive values for different arrhythmias) are not explicitly provided in a table within the provided text.

2. Sample size used for the test set and the data provenance:

  • Test Set Sample Size: "B-Secur evaluated HeartKey Rhythm against a large independent proprietary set of ECGs and standard databases." It specifies a proprietary dataset including more than 1200 ECGs.
  • Data Provenance:
    • Country of Origin: Approximately 66% of ECGs recorded in the US. The origin of the remaining 34% is not specified but implied to be non-US or unknown.
    • Retrospective or Prospective: Not explicitly stated, but "recorded ECGs" suggests a retrospective dataset.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

The document mentions that the device's analysis output was compared "against a known reference." It does not specify the number of experts used or their qualifications (e.g., radiologist with 10 years of experience) for establishing this ground truth.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

The document does not specify the adjudication method used for the test set's ground truth.

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:

A multi-reader multi-case (MRMC) comparative effectiveness study comparing human readers with and without AI assistance was not explicitly described in the provided text. The performance testing appears to be a standalone (algorithm only) evaluation against ground truth.

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

Yes, a standalone (algorithm only) performance evaluation was done. The text states: "B-Secur evaluated HeartKey Rhythm against a large independent proprietary set of ECGs and standard databases, comparing its analysis output against a known reference." This indicates the algorithm's performance was assessed independently.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

The ground truth was established by "a known reference," and the context of evaluating ECG analysis suggests this refers to expert consensus interpretation of ECGs, although the specifics of how this consensus was reached (e.g., multiple experts, adjudication) are not detailed.

8. The sample size for the training set:

The document does not specify the sample size for the training set used for the AI/ML model. It only describes the test set.

9. How the ground truth for the training set was established:

The document does not specify how the ground truth for the training set was established.

§ 870.1425 Programmable diagnostic computer.

(a)
Identification. A programmable diagnostic computer is a device that can be programmed to compute various physiologic or blood flow parameters based on the output from one or more electrodes, transducers, or measuring devices; this device includes any associated commercially supplied programs.(b)
Classification. Class II (performance standards).