(151 days)
The ZywieAI Software Library is intended for use by qualified medical professionals for the assessment of arrhythmias using historic ambulatory ECG data. The product supports downloading and analyzing data recorded in compatible formats from any device used for the arrhythmia diagnostics such as Holter, Event Monitor, MCT (Mobile Cardio Telemetry), or other similar devices when assessment of the rhythm is necessary. The ZywieAI Software Library can also be electronically interfaced, and perform analysis with data transferred from other computer based ECG systems, such as ECG management system. The software library provides ECG signal processing and analysis on a beat by beat basis, QRS Detection, Non-paced Arrhythmia Interpretation, PAC (Premature Atrial Contraction), Non-paced Heart Rate determination, Pause, Tachycardia, Bradycardia, Atrial Flutter/Fibrillation, Ventricular Flutter/Fibrillation, Ventricular Ectopic Beat detection, Ventricular Tachycardia, AV (Atrioventricular) Conduction Block, ST Change Episode detection, and Non-paced Ventricular Arrhythmia calls.
The product can be integrated into computerized ECG monitoring devices. In this case the medical device manufacturer will identify the indication for use depending on the application of their device.
The product cannot be used with potentially life-threatening arrhythmias which require inpatient monitoring.
Verification of the output from ZywieAI Software is the responsibility of a trained healthcare professional or physician.
The ZywieAI Software Library is used for analyzing ECG data for adult patient population.
The ZywieAI Software Library is an "object library". An object library is a collection of callable functions that have been compiled (or assembled) into machine code or Interactive Data Language (IDL) code for the computer on which they execute.
The basic software application reads the input ECG signals from a file into computer memory and passes that data to ZywieAI Software Library for analyzing, annotating, and creating output. This output is sent back to the basic software application where the output is written to a file. The read input signal may invoke some or all of the functions in the object library. An application program could be written to write input data to a file using web services. The same or a different application program could be written to consume the output file using web services.
The ZywieAI Software Library provides ECG signal processing, ORS Detection, Nonpaced Arrhythmia Interpretation, PAC, Non-paced Heart Rate determination, Pause, Tachycardia, Bradycardia, Atrial Flutter/Fibrillation, Ventricular Flutter/Fibrillation, Ventricular Ectopic Beat detection, Ventricular Tachycardia, AV Conduction Block, ST Change Episodes detection, Non-paced Ventricular Arrhythmia calls and rhythm interpretation.
The library can be accessed through an Application Program Interface (API) as a callable function or using a web service call. This allows the library to be used as an accessory to an ECG management application or as a stand-alone product.
Zywie will compile the ZywieAI Software Library as specified by an ECG device manufacturer. An object library will be created and delivered to the device manufacturer. who can then integrate it into application software for their ECG analysis.
This document describes the ZywieAI Software Library, an automated ECG analysis and interpretation software. Here's an analysis of the acceptance criteria and study information provided:
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't explicitly present a table of acceptance criteria with specific thresholds for performance metrics. Instead, it states that the ZywieAI Software Library "meets the requirements of following performance standards" and provides the names of those standards. The performance "reported" is that the device meets these standards.
| Performance Standard | Acceptance Criteria (Implied) | Reported Device Performance |
|---|---|---|
| ANSI/AAMI EC57:20012 Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms | Compliance with the methods and reporting requirements outlined in ANSI/AAMI EC57:20012 for rhythm and ST segment measurement algorithms. | The ZywieAI Software Library meets the requirements of ANSI/AAMI EC57:20012. |
| IEC 62304:2006 Medical device software Software lifecycle processes | Compliance with the software lifecycle processes, including quality, risk management, and documentation, as defined in IEC 62304:2006. | The ZywieAI Software Library meets the requirements of IEC 62304:2006. |
2. Sample Size Used for the Test Set and Data Provenance:
The document implicitly refers to test sets derived from standard public databases:
- Test Set Databases: MITDB (The MIT-BIH Arrhythmia Database), NSTDB (The MIT-BIH Noise Stress Test Database), QTDB (The QT Database - PhysioNet), & ESCDB (The European ST-T Database - PhysioNet).
- Sample Size: The exact sample size from these databases used for testing is not explicitly stated.
- Data Provenance: The databases mentioned (MIT-BIH, PhysioNet) generally consist of retrospective ECG recordings. The countries of origin for these specific datasets are typically the United States (MIT-BIH) and European institutions (ESCDB), but the document does not specify.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts:
The document does not explicitly state the number or qualifications of experts used to establish the ground truth for the test sets from the mentioned databases. However, standard databases like MIT-BIH and PhysioNet databases typically have their ground truth annotations established by multiple, qualified cardiologists or electrophysiologists.
4. Adjudication Method for the Test Set:
The document does not describe an explicit adjudication method for the test set. For publicly available and widely used databases like those mentioned, the ground truth annotations were established during the creation of the databases themselves, often through expert review and consensus, but the specific adjudication method for their use in this study is not detailed.
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:
The document does not mention an MRMC comparative effectiveness study involving human readers. The validation focuses on the standalone performance of the ZywieAI Software Library against established standards and existing predicate devices. The study performed is not designed to measure the improvement of human readers with AI assistance.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done:
Yes, a standalone study was done. The entire validation reported focuses on the performance of the "ZywieAI Software Library" itself, as an "object library" that processes ECG signals. The statement "Verification of the output from ZywieAI Software is the responsibility of a trained healthcare professional or physician" emphasizes that the software provides analysis, but the final interpretation and clinical decision are with a human, indicating the software's role as a standalone analytical tool.
7. The Type of Ground Truth Used:
The ground truth used for testing was derived from expert-annotated databases (MIT-BIH Arrhythmia Database, NSTDB, QTDB, ESCDB). These databases are renowned for having meticulously annotated ECG recordings, often by cardiologists or electrophysiologists, which serve as the reference standard for various arrhythmias and ECG events.
8. The Sample Size for the Training Set:
The document does not explicitly state the sample size, composition, or provenance of the training set used for the ZywieAI Software Library. It only discusses the data used for testing/verification.
9. How the Ground Truth for the Training Set Was Established:
The document does not provide any information on how the ground truth for the training set was established. As common with medical device submissions, details about internal training data and their ground truth establishment are often proprietary and not disclosed in the public 510(k) summary.
§ 870.2340 Electrocardiograph.
(a)
Identification. An electrocardiograph is a device used to process the electrical signal transmitted through two or more electrocardiograph electrodes and to produce a visual display of the electrical signal produced by the heart.(b)
Classification. Class II (performance standards).