Search Results

The MR Workstations (including the ViStar, Montage and OnCall) are designed to provide fast and convenient review of DICOM compatible digital images from all six diagnostic imaging modalities (MR, CT, Nuclear Medicine, X-Ray, CR and Ultrasound) and secondary capture as well as post-processing of MR and CT images including multi planar reformatting, angiographic rendering, surface rendering, segmentation/disarticulation and cine display.

When used in conjunction with any of these diagnostic imaging modalities, the indications for use remain unchanged from those defined for that modality.

The Picker MR Workstations (ViStar, Montage and OnCall) integrate image review and processing software with a variety of hardware configurations. The software consists of a graphical user interface and post-processing capabilities for MR and CT images including: multi planar reformatting, angiographic rendering, surface rendering, segmentation / disarticulation and cine display. The hardware configurations can be divided into three major types, ViStar, Montage and Oncall. In general, the ViStar line of products are single-monitor workstations which can be equipped with all of the image manipulation and reformating capabilities described above. The Montage line of products are multi-monitor, multimodality viewing stations which have limited image manipulation and reformating capabilities. Finally, the OnCall systems are PC-based for remote viewing and provide image review and cine display.

The provided text does not describe a study that proves the device meets specific acceptance criteria in the manner usually associated with clinical validation studies for AI/medical devices. Instead, it relies on a Substantial Equivalence Chart to demonstrate that the MR Workstations (ViStar, Montage, and OnCall) are substantially equivalent to previously cleared predicate devices (ViStar K905070 and EDGE/VISTA K931544).

This document is from 1996, a time when AI in medical imaging was not yet a prominent field, and validation methods were different. The "device" in question here is an image viewing and processing workstation, not an AI algorithm. Therefore, the questions related to AI-specific validation (like standalone performance, MRMC studies for AI improvement, and sample sizes for training/test sets for AI models) are not directly applicable.

However, I can extract the "acceptance criteria" and "reported device performance" based on the substantial equivalence comparison. In this context, "acceptance criteria" can be interpreted as the characteristics and functionalities of the predicate devices, and "reported device performance" are the characteristics of the new device that align with or exceed those of the predicate.

Here's the information formatted as requested, with notes explaining the lack of certain AI-specific details:

Acceptance Criteria and Reported Device Performance

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

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

• Not Applicable. This document pre-dates AI-specific validation studies involving test sets derived from patient data. The "study" here is a comparison of technical specifications and intended uses for substantial equivalence, not a performance evaluation on a patient dataset.

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

• Not Applicable. No ground truth for a test set of medical images was established as part of this submission for substantial equivalence.

4. Adjudication method for the test set:

• Not Applicable. No test set or ground truth adjudication was involved.

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. This is not an AI device. No MRMC study was performed comparing human performance with or without AI assistance.

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

• Not Applicable. This is an image workstation, not an AI algorithm designed for standalone performance.

7. The type of ground truth used:

• Not Applicable. No ground truth data was used in the context of an algorithm's performance. The "ground truth" for this submission is implicitly the established functionality and safety of the predicate devices, against which the new device's specifications are compared.

8. The sample size for the training set:

• Not Applicable. This device is an image processing workstation, not an AI model that requires a training set.

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

• Not Applicable. There was no training set for an AI model.

Ask a specific question about this device