(90 days)
The Vascular Analysis Package is intended for off-line image processing and manipulation of digitally acquired datasets from Integris systems. The use of the Vascular Analysis package is comparable with that on the Integris operator console (the algorithms used with the Easyvision are identical to those already used on the Integris systems, although the user interface is different).
With the Vascular Analysis package the user can optimize images and perceived image quality for vascular application specific purposes.
The vascular analysis package supports the postprocessing on Digital Subtraction Angiography (DSA) images done to perform diagnosis. In a vascular study series of images are acquired either with or without contrast media. These contrast media can be iodine or CO2, and they make the vasculature visible in the image. The features in the vascular package allow for optimizing these images in the following way. Subtraction is a common used technique in DSA . Two images representing the same part of the anatomy (one with contrast media and one without contrast media) are subtracted from each other. The resulting image is an image that displays the vascular anatomy/pathology only, due to the fact that all equal anatomical structures in the two images are subtracted from each other. Thereby leaving only visible the vascular structures filled with contrast media. Run subtraction is identical to subtraction but now the subtraction process is applied to two series of images wherein image 1 of series 1 correlates w.r.t anatomy/position to image 1 in series 2.
Landmarking will allow to bring back some of the anatomical structure back into the image as to relate vascular pathology to anatomical position. Sometimes there can be a mismatch between two images that are subtracted from each other due to patient movement. To correct for this movement pixel shift is applied to the image pair, where the mask image (image without the contrast) is shifted in horizontal and/or vertical axis compared to the contrast image. This will enhance the visibility of the vascular structure only.
This can either be done:
- Manually, the user shifts in a Region Of Interest (ROI) to have an optimal result. The . shift result in the ROI will be applied to the whole image.
- Split Screen, to allow different shifts in one subtracted pair. The image is either divided . by a horizontal or vertical line into two areas. In each of those two areas different values of shift can be applied.
- . AutoWarp, a means of enhancing a pair of subtracted images by adapting regions in the mask image to regions in the contrast image.
The provided 510(k) summary for the Philips Easy Vision Family Workstation Option, Vascular Analysis, states that "The algorithms used with the Easyvision are identical to those already used on the Integris systems, although the user interface is different." This implies that the device's acceptable performance is benchmarked against the predicate Integris systems, and the FDA's clearance is based on this substantial equivalence. However, the document does not describe specific acceptance criteria or an independent study detailing the device's performance against such criteria for the new workstation.
The 510(k) summary focuses on demonstrating that the Easy Vision Vascular Analysis package is substantially equivalent to the predicate Philips Integris systems and Siemens systems. It describes the intended use and system description, detailing the image processing and manipulation functions for Digital Subtraction Angiography (DSA) images, such as subtraction, run subtraction, landmarking, and pixel shift methods (manual, split screen, and AutoWarp).
Without a dedicated performance study for the Easy Vision workstation or explicitly stated acceptance criteria in the provided text, the specific details requested in the prompt cannot be fully extracted for this particular device. The information below reflects what can be inferred or is directly stated regarding the predicate device's performance and the basis for substantial equivalence, rather than a separate study for the Easy Vision workstation itself.
Here's an attempt to answer the questions based on the available information, noting the gaps:
1. A table of acceptance criteria and the reported device performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Functional Equivalence to Predicate Device: The algorithms used for image processing and manipulation (subtraction, run subtraction, landmarking, pixel shift, AutoWarp) for Digital Subtraction Angiography (DSA) should be identical to those on the Integris systems. | The document explicitly states: "the algorithms used with the Easyvision are identical to those already used on the Integris systems, although the user interface is different." This statement serves as the primary evidence for meeting the functional equivalence criterion. |
| Image Quality Optimization: The device should allow users to optimize images and perceived image quality for vascular application-specific purposes. | The system description details how various features (subtraction, landmarking, pixel shift) contribute to optimizing images by enhancing visibility of vascular structures and correcting for patient movement. The substantial equivalence argument implies that the image quality optimization capabilities are comparable to the predicate. |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Sample Size: Not specified for a dedicated test set for the Easy Vision workstation. The claim of "identical algorithms" suggests reliance on prior validation of the Integris systems.
- Data Provenance: Not specified.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
- Not specified. Given the reliance on substantial equivalence, any prior validation on the Integris likely involved expert review, but details are not provided for the Easy Vision.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Not specified.
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 specified. This is a workstation that provides tools for image manipulation, rather than an AI-assisted diagnostic tool in the modern sense. The "AI" (AutoWarp) mentioned is a specific image processing algorithm.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- While the algorithms themselves function without human interaction once initiated (e.g., AutoWarp), the document describes a "Workstation" and "Vascular Analysis package" where a user ("the user can optimize images") interacts with the system. The focus is on the tools available to the user for post-processing, implying human-in-the-loop performance is inherent to its intended use. No standalone evaluation of the algorithms separate from the workstation in a user context is described.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- Not specified. For a system processing DSA images, the "ground truth" for evaluating image optimization would typically involve expert assessment of image quality, clarity of vascular structures, and accuracy of anatomical representation after processing, likely against clinical or expert-interpreted images.
8. The sample size for the training set
- Not applicable as this document focuses on a 510(k) submission based on substantial equivalence, not a de novo algorithm development and training. The algorithms are stated to be "identical to those already used on the Integris systems," implying they were previously developed and validated.
9. How the ground truth for the training set was established
- Not applicable for the same reasons as above.
Summary of Gaps and Inferences:
The provided 510(k) summary primarily asserts substantial equivalence to predicate devices (Philips Integris, Siemens) by stating that the algorithms used in the Easy Vision Vascular Analysis package are "identical" to those already in use on the Integris systems. This means that the regulatory acceptance is built upon the prior demonstrated safety and effectiveness of the predicate devices. The document does not detail new, independent performance studies with specific acceptance criteria, sample sizes, expert qualifications, or ground truth establishment for the Easy Vision workstation itself. Instead, the "study" proving the device meets acceptance criteria is essentially the assertion and FDA's acceptance of the "identical algorithms" claim.
§ 892.2050 Medical image management and processing system.
(a)
Identification. A medical image management and processing system is a device that provides one or more capabilities relating to the review and digital processing of medical images for the purposes of interpretation by a trained practitioner of disease detection, diagnosis, or patient management. The software components may provide advanced or complex image processing functions for image manipulation, enhancement, or quantification that are intended for use in the interpretation and analysis of medical images. Advanced image manipulation functions may include image segmentation, multimodality image registration, or 3D visualization. Complex quantitative functions may include semi-automated measurements or time-series measurements.(b)
Classification. Class II (special controls; voluntary standards—Digital Imaging and Communications in Medicine (DICOM) Std., Joint Photographic Experts Group (JPEG) Std., Society of Motion Picture and Television Engineers (SMPTE) Test Pattern).