(65 days)
The Medtronic CardioInsight® Cardiac Mapping System is intended for acquisition, analysis, display and storage of cardiac electrophysiological data and maps for analysis by a physician.
The Medtronic CardioInsight® Cardiac Mapping System is a non-invasive mapping system for beat-by-beat, multichamber, 3D mapping of the heart. The CardioInsight Cardiac Mapping System displays cardiac maps and virtual electrograms from real-time chest ECG signals (measured by a Sensor Array placed on the torso) and CT scan data. The ECG signals in concert with the CT scan information (geometrical information) are used in mathematical algorithms to transform the measured body surface signals into epicardial signals via solving the cardiac inverse problem. The CardioInsight Cardiac Mapping System software uses this data to provide various cardiac signal analyses and displays interactive 3D color maps including potential, activation, voltage, propagation, and phase maps. The CardioInsight Cardiac Mapping System system is mobile and can be used for mapping at the patient's bedside or in the EP lab.
This looks like a 510(k) summary for the Medtronic CardioInsight® Cardiac Mapping System.
Based on the provided document, the device described is a cardiac mapping system, not an AI/ML algorithm that predicts or diagnoses. Therefore, much of the requested information (like acceptance criteria for AI performance, sample sizes for training/test sets, expert adjudication, MRMC studies, or standalone algorithm performance) is not applicable to this device submission.
The document focuses on demonstrating substantial equivalence to a predicate device (ECVUE MAPPING SYSTEM) based on shared intended use, technology, and performance, rather than providing specific performance metrics against clinical acceptance criteria in the context of an AI model's output.
Here's what can be extracted and what is explicitly stated as not applicable:
1. Table of Acceptance Criteria and Reported Device Performance:
The document describes various performance tests conducted to verify the device's compliance with safety and specifications, and that it "performs as designed." However, it does not present a table of specific clinical acceptance criteria with corresponding performance metrics in the way one would for a diagnostic or predictive AI device. The performance data highlighted are primarily related to electrical safety, mechanical safety, software/firmware verification, algorithm testing (to ensure it met requirements and functioned as intended), packaging validation, and usability testing.
| Performance Area | Acceptance Criteria (Implied/General) | Reported Device Performance |
|---|---|---|
| Overall Performance | Suitable for its intended use, complies with safety and specifications, performs as designed. (General statement, specific quantitative criteria for mapping accuracy are not provided in this summary.) | "Performance testing was completed on the CardioInsight Cardiac Mapping System which verified that the System complies with the safety and specifications and performs as designed; it is suitable for its intended use." |
| "Testing demonstrated that the CardioInsight Cardiac Mapping System met the requirements and functioned as intended." | ||
| Hardware Verification | Compliance with dimensions, cart functionality, mechanical safety, amplifier mechanical and electrical functionality, and electrical safety. | Verified. |
| Labeling Verification | Compliance with applicable requirements. | Verified. |
| Electrical Safety | Compliance per ANSI/AAMI ES 60601-1:2005/A1:2012, IEC 60601-1:2005/A1:2012, EN 60601 1:2006/A1:2013 (excluding biocompatibility, usability, EMC). | Compliant. |
| EMC/EMI | Compliance per AAMI / ANSI / IEC 60601-1-2:2007/(R)2012. | Compliant. |
| Mechanical Safety | Compliance per ANSI/AAMI ES 60601-1:2005/A1:2012, IEC 60601-1:2005/A1:2012, EN 60601 1:2006/A1:2013 (excluding biocompatibility, usability, EMC). | Compliant. |
| Software/Firmware/Algorithm | Software: Met requirements per "FDA's Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices" and AAMI / ANSI / IEC 62304:2006 (moderate level of concern). Firmware: Met requirements. Algorithms: Met requirements and functioned as intended, and performed as expected when integrated. | Verified. Demonstrated firmware met requirements. Algorithms met requirements and functioned as intended and performed as expected when integrated. |
| Packaging Validation | Met environmental conditioning and simulated shipping per applicable sections of ASTM D4332-14 and ASTM D4169. | Demonstrated. |
| Usability Testing | Performed. | Performed. |
| System V&V (Functionality & Performance) | Performed in a simulated environment. | Performed. |
2. Sample size used for the test set and the data provenance:
- Not Applicable in the traditional AI sense. The document states: "Preclinical and clinical testing were not required for the CardioInsight Cardiac Mapping System." This implies there wasn't a specific "test set" of clinical cases used to evaluate an algorithm's performance against ground truth in the way an AI model would be tested. The validation was primarily system functionality and safety.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not Applicable. As no clinical "test set" was described for an AI algorithm's output, there's no mention of experts establishing ground truth in this context.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Not Applicable.
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:
- Not Applicable. This is not an AI-assisted diagnostic device where human reader improvement would be measured. The device provides data and maps for analysis by a physician.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not Applicable. The device is a system for "acquisition, analysis, display and storage of cardiac electrophysiological data and maps for analysis by a physician." It's an instrument providing information to a human. There is no mention of a standalone diagnostic algorithm being evaluated.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not Applicable in the context of AI performance. The system's "truth" is its ability to accurately acquire and transform ECG and CT data into epicardial signals and maps, which is verified through engineering and functional testing rather than clinical expert ground truth for interpretation.
8. The sample size for the training set:
- Not Applicable. The device is not an AI/ML model that undergoes a "training phase" with a specific dataset. Its algorithms are based on mathematical principles (solving the cardiac inverse problem) and are validated through standard software and system verification.
9. How the ground truth for the training set was established:
- Not Applicable. See point 8.
§ 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).