Search Results

The GE EchoPAC BT10 workstation is indicated for diagnostic review and analysis of ultrasound images acquired under various modes of operation including B, M, Color M modes, Color, Power, Pulsed & CW Doppler modes, Coded Pulse, Harmonic and Realtime 3D. Clinical applications include: Fetal; Abdominal; Urology (including prostate); Pediatric; Small Organ (breast, testes, thyroid); Neonatal and Adult Cephalio; Cardiac (adult and pediatric); Peripheral Vascular; Transesophageal (TEE); Musculo-Conventional; Transrectal (TR); Transvaginal (TV); and Intraoperative skeletal (abdominal, thoracic, & vascular).

GE EchoPAC provides image processing, annotation, analysis, measurement, report generation, communication, storage and retrieval of ultrasound images that are acquired via GE Vivid family of ultrasound scanners, primarily for cardiology ultrasound applications but also for general imaging. The EchoPAC software is an integral component of each Vivid system, providing the post acquisition image management and reporting functions of the scanner. Sold as a stand-alone software only product it can be installation on the customer's PC hardware, or as a plug-in to third party PACS. EchoPAC is DICOM compliant, transferring images and data via LAN between scanners, hard copy devices, file servers and other workstations. The modified or added software features for GE EchoPAC BT10 are substantially equivalent to the unmodified device and functionality cleared on GE Vivid E9 and GE Vivid S5/S6.

This document, K101324, describes a 510(k) Premarket Notification for the GE EchoPAC BT10 Review station. It's important to note that this is a submission for a review station software and not a diagnostic AI device in the modern sense. Therefore, many of the performance metrics and study designs typically associated with AI/ML diagnostic tools (like sensitivity, specificity, MRMC studies, training set details) are not applicable here. The submission focuses on demonstrating substantial equivalence to predicate devices, primarily through non-clinical testing and verification of design specifications.

Here's an analysis based on the provided text, addressing your questions to the extent possible:

Acceptance Criteria and Device Performance:

Since this is not a diagnostic AI device, the "acceptance criteria" are not framed in terms of clinical performance metrics like sensitivity or specificity. Instead, they are focused on design specifications, compliance with standards, and functional equivalence to predicate devices.

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

• Not applicable / Not explicitly stated. This submission focuses on non-clinical testing and verification. There is no mention of a "test set" in the context of clinical data or patient samples being analyzed for performance metrics. The testing would have involved software validation and verification against functional requirements, not clinical diagnostic accuracy.

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

• Not applicable. As there was no clinical "test set" in the diagnostic sense, there was no need for experts to establish ground truth for such a set.

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

• Not applicable. No clinical test set requiring adjudication was used.

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:

• No. An MRMC study was not conducted. This device is a review station, not an AI-powered diagnostic tool, and its purpose is not to assist human readers in a diagnostic capacity that would be measured by an MRMC study.

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

• Not applicable. The GE EchoPAC BT10 is workstation software for image review, analysis, and reporting, which by its nature is a human-in-the-loop system. It is not an algorithm that operates standalone to produce diagnostic outputs.

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

• Not applicable / System Functionality, Compliance, and Equivalence. For this type of device, "ground truth" would relate to the correct functioning of the software, its adherence to design specifications, and its ability to process and display images as intended, consistent with DICOM standards and the functionality of predicate devices. There wouldn't be a clinical "ground truth" in the sense of a disease state.

8. The sample size for the training set:

• Not applicable. As the GE EchoPAC BT10 is not an AI/ML device that requires a "training set" in the machine learning context, this information is not relevant.

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

• Not applicable. There was no training set for a machine learning model.

Ask a specific question about this device