Search Results

STAR Apollo™ Mapping System assists users in manual annotation of 3D anatomical and electrical maps of human atria using data from multipolar, intracardiac, atrial, electrograms during atrial fibrillation. The clinical significance of utilizing the STAR Apollo Mapping System, to help identify areas with intracardiac atrial arrhythmias, such as atrial fibrillation, has not been established by clinical investigations.

The STAR Apollo 100 Mapping System is a software driven system designed to assist operators in identifying Early Site of Activations (ESA) and Repetitive Patterns of Activation (RPA) in patients undergoing a cardiac mapping procedure for Atrial Fibrillation (AF). The software is designed for use in association with a commercially approved electroanatomic mapping system, specifically, Ensite Precision Cardiac Mapping System Model EE 3000 (V2.6) or Ensite X EP System (V 1.1.1) (Abbott Medical), and the commercially approved catheter, Advisor™ HD Grid Mapping Catheter, Sensor Enabled™ (Abbott Medical). The system consists of proprietary Star Apollo Mapping System software and a hardware component. Star Apollo Mapping System software consists of 3 main components: Electroanatomic data import, the STAR Apollo Mapping System engine (C++ code) and Graphics User Interface (GUI). The STAR Apollo Mapping System is designed to run on a laptop computer running Windows 10 or later operating system. Star Apollo Mapping System software is pre-installed onto the laptop. The STAR Apollo Mapping System uses export data from the Ensite Precision or Ensite X EP system that has been collected with the HD Grid catheter during the mapping procedure. The HD Grid is used to collect anatomy, localization and electrogram data in the atria. Recordings are made for 30 seconds with the HD Grid in a stable position and in contact with the atrial wall. These 30 seconds acquisitions are made in multiple, non-overlapping locations, so as to generate recordings over the entire atrial chamber. Once the data has been acquired, the data is anonymized and exported, via a portable external data storage device, to the laptop computer running the STAR Apollo Mapping System. The export data from the Ensite Precision or Ensite X EP systems consists of electrograms, electrode coordinates, ECG recordings and the geometry model. The data are imported, utilizing the portable external data storage device, into the STAR Apollo Mapping System and then processed by the STAR Apollo engine to generate a STAR Apollo map visualized by the GUI.

The provided FDA 510(k) summary for the STAR Apollo™ Mapping System outlines the device's acceptance criteria and the studies conducted to demonstrate its performance.

Here's an analysis of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are primarily derived from "Software design verification testing" and "Verification of algorithm calculations and graphic output." The performance is reported by stating that the software met these criteria.

2. Sample Sizes and Data Provenance

• Test Set Sample Size: Not explicitly stated in the provided document. The descriptions refer to "test cases" and "electrophysiologists performing manual annotation of ECGs" and "electrophysiologists and rated the overall use." No specific number of cases or patient data points is given.
• Data Provenance: Not explicitly stated. The studies are described as "simulated testing." It is not clear if these simulations used real patient data (retrospective/prospective) or synthetically generated data. There is no mention of country of origin for the data.

3. Number of Experts and Qualifications for Ground Truth

• Number of Experts: Not explicitly stated. The document mentions "electrophysiologists" (plural), but not a specific count.
• Qualifications of Experts: They are identified as "electrophysiologists," which implies medical doctors specializing in cardiac electrophysiology. No further details regarding their experience level (e.g., years of experience, board certification) are provided.

4. Adjudication Method for the Test Set

• Adjudication Method: Not explicitly stated. For the "Blinded Physician Validation," it says "electrophysiologists performing manual annotation of ECGs and comparing the results with the STAR Apollo Mapping System." This suggests a comparison to individual expert annotation rather than a formal consensus or adjudication process among multiple experts to establish a single ground truth.

5. Multi-Reader, Multi-Case (MRMC) Comparative Effectiveness Study

• MRMC Study: No, a formal MRMC comparative effectiveness study demonstrating how human readers improve with AI vs. without AI assistance was not reported. The studies described are:
  • Blinded Physician Validation: This compares the device's output to manual human annotation, suggesting a standalone performance aspect relative to human interpretation, but not the human-in-the-loop improvement.
  • STAR Apollo Physician Use: This assessed the usability and appropriateness of the device, which is a qualitative assessment of user experience, not an effectiveness study.
• Effect Size: Not applicable, as no MRMC study was performed or reported.

6. Standalone (Algorithm Only) Performance

• Standalone Performance: Yes, aspects of standalone performance were evaluated.
  • "Algorithm description and testing" and "Verification of algorithm calculations and graphic output" describe testing the core algorithms' accuracy against predetermined outputs, which is a standalone assessment.
  • The "Blinded Physician Validation (Manual Annotation versus STAR Apollo Mapping System)" also assesses the device's output independently, comparing it to human annotation, rather than measuring human improvement when using the device.

7. Type of Ground Truth Used

• Ground Truth Type:
  • For algorithm and software verification, the ground truth was "correct, predetermined output" and "acceptance criteria for each of the test cases" set during development. This implies a ground truth defined by the developers based on internal specifications.
  • For the "Blinded Physician Validation," the ground truth was "manual annotation" performed by electrophysiologists, essentially using expert interpretation as a reference. There is no mention of pathology or outcomes data as ground truth.

8. Sample Size for the Training Set

• Training Set Sample Size: Not explicitly stated. The document describes the device as using "fixed algorithms and equations" rather than deep learning or machine learning that would typically require a training set. The comparison to the predicate (VX1) explicitly states: "VX1 is a software that utilizes deep learning based algorithms to analyze data sets, whereas the STAR Apollo Mapping System software uses fixed algorithms and equations to analyze the data sets." Therefore, a traditional "training set" as understood in machine learning may not apply in the same way, or its size is not relevant for a fixed-algorithm system.

9. How Ground Truth for Training Set was Established

• Training Set Ground Truth Establishment: Not applicable given that the device uses "fixed algorithms and equations" and not deep learning, negating the need for a labeled training dataset in the conventional sense. The "ground truth" for its development would be the fundamental principles of electrophysiology and the expected mapping outcomes that the algorithms are designed to mimic.

Ask a specific question about this device