(23 days)
The Siemens Symbia series is intended for use by appropriately trained health care professionals to aid in detecting, localizing, diagnosing, staging and restaging of lesions, tumors, disease and organ function for the evaluation of diseases and disorders such as, but not limited to, cardiovascular disease, neurological disorders and cancer. The images produced by the system can also be used by the physician to aid in radiotherapy treatment planning and interventional radiology procedures.
Software: The MI Applications software is a display and analysis package intended to aid the clinician in the assessment and quantification of pathologies taken from SPECT, PET, CT and other imaging modalities.
Symbia.net introduces client/server functionality allowing the MI Applications product to be deployed on any compatible hardware (including desktops, laptops and workstations) that meet minimal hardware requirements. Symbia.net is a solution for SPECT. SPECT-CT. PET. and PET-CT systems.
The software application is provided on a server. Clients running on any personal computer or Mac, meeting minimum requirements, can have access to the system. The rendered images on the client and server side are clinically equivalent with each other. No image quality deqradation on any of the images shown in the clients compared to the images shown on the server related to resolution, color and timing.
Here's an analysis of the provided text regarding the acceptance criteria and supporting study for the Symbia.net VA10B device.
Important Note: The provided document is a 510(k) Summary and FDA Clearance Letter. These documents primarily focus on demonstrating substantial equivalence to predicate devices and confirming regulatory compliance, rather than detailing the specifics of performance studies in the same way a clinical trial report would. Therefore, much of the requested information regarding sample sizes, ground truth establishment, expert qualifications, and MRMC studies is not explicitly present in this type of regulatory filing.
However, based on the principle of substantial equivalence, the performance of Symbia.net VA10B is implicitly accepted as equivalent to its predicate devices. The study performed focuses on demonstrating that the new client-server architecture does not degrade image quality compared to the established predicate system.
Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Image Quality: No image quality degradation related to resolution, color, and timing when rendered on client-side compared to server-side. | "The rendered images on the client and server side are clinically equivalent with each other. No image quality degradation on any of the images shown in the clients compared to the images shown on the server related to resolution, color and timing." |
Study Details (Based on available information)
-
Sample size used for the test set and the data provenance:
- Test set sample size: Not explicitly stated in the provided document. The study likely involved a set of images or image series that were rendered on both the server and client.
- Data provenance: Not specified (e.g., country of origin, retrospective/prospective). Given this is a software update for an existing system, it's highly probable that existing clinical image data was used, but details are not provided.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Number of experts: Not explicitly stated.
- Qualifications of experts: Not explicitly stated. However, for a "clinical equivalence" assessment in medical imaging, it is standard practice to involve qualified medical professionals (e.g., radiologists, nuclear medicine physicians) to evaluate image quality and interpret findings.
-
Adjudication method for the test set:
- Not explicitly stated. Given the nature of comparing rendered images for degradation, it would likely involve visual comparison by experts. Common methods include blinded review, side-by-side comparison, or comparing quantitative metrics if applicable, but the specific adjudication method is not described.
-
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 comparative effectiveness study was not explicitly done or mentioned. This submission is for a client/server deployment of an existing MI Applications product, not a new AI-based diagnostic tool. The focus is on ensuring the new deployment method does not degrade the established performance. Therefore, there's no "AI assistance" component to measure improvement from.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, in spirit. The core of the performance demonstration is a technical validation that directly compares the output of the software (rendered images) in two different deployment configurations (server vs. client). While humans are undoubtedly involved in assessing the equivalence, the "algorithm" (i.e., the rendering and display functionality) is being evaluated for its standalone output characteristics (resolution, color, timing) across the two environments. No human interaction is part of the core algorithm's performance in this context, rather the display output of the software itself is being assessed.
-
The type of ground truth used:
- Expert Consensus/Visual Inspection and Technical Comparison. The "ground truth" here is the original image quality and clinical appearance as displayed on the server. The study aims to demonstrate that the client display maintains this original quality. This is typically assessed by experts visually confirming clinical equivalence and potentially through technical measurements of image parameters (resolution, color fidelity, timing).
-
The sample size for the training set:
- Not applicable/Not explicitly stated. This device is a Picture Archiving and Communication System (PACS) and image display/analysis software. It's not described as using machine learning or AI models that require a "training set" in the conventional sense. The "training" would be more akin to software development and testing cycles using various image types, rather than a data-driven machine learning training set.
-
How the ground truth for the training set was established:
- Not applicable/Not explicitly stated. As there is no mention of a traditional machine learning training set, the concept of establishing ground truth for it is not relevant to the provided text. The "truth" for the development of such a system resides in ensuring the software correctly processes, renders, and displays medical images according to established DICOM standards and clinical requirements.
§ 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).