(29 days)
Merge PACS™ is a Picture Archiving and Communication System (PACS) for multi-modality (CT, MR, PT, US, MG, BTO, CR, DR/DX, NM, XA, RF, secondary capture (SC), and other DICOM-compliant modalities) image processing and display, diagnostic reading and reporting, communication, printing, and storage of medical imaging studies and other patient data.
Intended clinical users include radiologists, orthopedic and other surgeons, referring physicians, technologists, and other qualified medical professionals. Data can be received directly from acquisition modalities, CAD systems, and other image processing systems, or indirectly via importing. Data that is not DICOM-compliant, such as photos, can be converted to DICOM format by Merge PACS.
Merge PACS provides image manipulation tools to enable users to view and compare images such as: linkinq, MPR, MIP, 3D image fusion/registration of CT, MR, and PET; as well as CVR (Color Volume Rendering), measurements (linear distances, angles, areas, SUV, etc.), and annotations (for example, outline and label regions of interest, label spinal vertebrae).
The Real Time Worklist (RTWL) displays the real-time status of radiology activity and provides customizable workflow management capabilities. Communication of critical results is facilitated and documented through optional, configurable components.
The Patient Dashboard provides a composite view of patient data, both imaging and non-imaging. The optional Reach component provides clinicians with secure, proactive communication and access to clinical reports and images. Multi-tier patient identity matching provides a comprehensive view even when dealing with multiple disparate patient identities.
Order and report information generated by the HIS/RIS and report creation systems are received and displayed via the transmission of HL7 messaging. Lossless (reversible) and lossy (irreversible) image compression are supported for viewing, storage and communication.
Merge PACS displays full fidelity DICOM images for use in the diagnostic interpretation of mammography using MG or BTO images. Thick slab MIP presentation can be applied to BTO images.
Lossy compressed images and digitized screen film images must not be used for primary diagnosis of mammography studies, and only display monitors that have regulatory clearance for mammography interpretation should be used for the interpretation of mammography studies.
Merge PACS™, a software medical device, is a standards-based medical imaging diagnostic workstation that serves as an adjunct to assist the clinician to view, read, and report their findings. Merge PACS processes and displays medical images from DICOM-compliant modalities. The device is designed to enable efficient workflows by maintaining clinicians' worklists and retrieving and managing studies for reading, reporting, communication, and storage.
Merge PACS software runs on off-the-shelf computer hardware and can be configured to operate standalone or to integrate with vendor-neutral imaging archives (VNAs) such as iConnect Enterprise Archive (iCEA) for image storage, and with radiological and hospital information systems (RIS and HIS) and medical record systems (EMR, EHR, etc.).
Merge PACS can be accessed from within the hospital or enterprise, or from remote locations via web-based access. Images viewed on mobile devices must not be used for diagnostic review.
The focus of this premarket notification is on the addition of the ability to "fuse" images for viewing (image fusion) and on the ability to measure Standardized Uptake Values (SUV) on PET (Positron Emission Tomography) images.
The provided text is a 510(k) Pre-market Notification for the Merge PACS™ system. It outlines the device's intended use, comparison to predicate devices, and performance data. However, it explicitly states that clinical studies were not required, meaning specific acceptance criteria with reported performance, sample sizes, and expert reviews as requested in your prompt were not applicable or performed for this submission.
The document focuses on demonstrating substantial equivalence to existing devices (AMICAS PACS 6.0 and Fujifilm Synapse PACS) by comparing features and technologies, rather than proving performance against predefined quantitative acceptance criteria through clinical trials.
Therefore, many of the requested details cannot be extracted from this document. Here's what can be provided based on the text:
Key Takeaway: The submission is based on demonstrating substantial equivalence through feature comparison and non-clinical testing, not on meeting specific quantitative acceptance criteria established via clinical studies.
Summary of Device Performance (Based on Non-Clinical Testing and Feature Comparison):
Since no clinical studies with specific acceptance criteria were conducted, the "performance" is described in terms of compliance and functional validation.
-
Acceptance Criteria and Reported Device Performance: Not applicable as defined by your prompt (no quantitative clinical acceptance criteria provided or met). The "acceptance" is based on functional verification through non-clinical testing and comparison to predicates.
Table (Hypothetical, based on functional claims, not quantitative performance):
| Feature/Function | Acceptance Criteria (Implied by equivalence) | Reported Device Performance (Non-Clinical Validation) |
|---|---|---|
| Image Fusion (PET/CT/MR) | Accurate overlay and registration of images in 2D and 3D. | Device performs as expected; original images always available; rigid transformation used. |
| SUV Calculation (PET) | Accurate calculation of SUV values from individual pixels or ROI. | Device performs as expected; meets RSNA/QIBA guidelines using DRO. |
| DICOM Compliance | Adherence to DICOM standards for image and data format. | Complies with DICOM standards. |
| Workflow Management (e.g., RTWL) | Efficient display and management of radiology activity. | Functions as expected; provides customizable workflow. |
| Image Manipulation Tools (MPR, MIP, CVR, measurements) | Correct application and display of manipulation tools. | Functions as expected. |
- Sample size for the test set and data provenance: Not specified for any quantitative testing that would establish "test set" performance. The non-clinical testing would have involved internal datasets, but details on size, origin, or retrospective/prospective nature are not provided.
- Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. No ground truth established by experts for performance evaluation was mentioned. The "ground truth" for the non-clinical tests would have been the expected functional output.
- Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable. No expert adjudication method was mentioned.
- 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 MRMC study was done. The device is a PACS system, not an AI for diagnostic assistance in the traditional sense that "improves human readers."
- If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: The device is a software workstation with various functionalities, not a standalone AI algorithm with a distinct "performance" metric that would be evaluated in isolation from human interaction. Its functions (like SUV calculation) are tools for human users. Non-clinical tests confirmed these functions performed "as expected," which is a form of standalone functional validation.
- The type of ground truth used (expert consensus, pathology, outcomes data, etc.): For the non-clinical tests, the "ground truth" was likely the expected outcome of direct functional tests (e.g., if you input X, the SUV calculation should yield Y; if you overlay two images, they should align based on DICOM coordinates). There was no mention of using clinical outcomes, pathology, or expert consensus as a ground truth for performance evaluation in the context of "acceptance criteria."
- The sample size for the training set: Not applicable. The document describes a PACS system which is a software medical device, not a machine learning or AI model that typically has a "training set." The development of the software would involve traditional software engineering and testing processes, not a dataset-driven training phase in the AI sense.
- How the ground truth for the training set was established: Not applicable, as there is no "training set" in the context of an AI/ML model for this device.
§ 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).