(81 days)
Vault is intended to act as a long term storage device (archive) for medical images.
Vault is a networked device that acts as a DICOM Storage Service Class Provider (SCP):
- Images can be sent to it via the DICOM protocol, from modalities or other devices that are sources of medical images.
- Images can subsequently be queried and retrieved via the DICOM protocol, to imaging workstations and other processing devices.
- Internally, Vault is optimized to provide secure storage for a very large number of images, and to provide rapid retrieval on demand. To that end, Vault includes high capacity storage media, and has implemented hierarchical storage management software to manage cache storage effectively.
Image Vault is a software package which may be marketed as a software only solution, as well as in conjunction with standard PC hardware. Image Vault is a PC-based, DICOMcompliant PACS device that is able to send & receive, display, store, and archive DICOM non-compressed and JPEG-compressed images. Images can be archived using standard magnetic tapes. It also has the ability to read/write a collection of images to/from CD-R media in a method defined by DICOM.
Images can be viewed using a simple display utility, for the purposes of properly identifying patients and their corresponding studies, and is not intended to be used for diagnosis.
Image Vault uses standard "off-the-shelf" PC hardware and communicates using the standard TCP/IP stack. The network hardware used to support the TCP/IP stack is superfluous to Image Vault.
The provided document is a 510(k) summary for the Mitra Imaging, Inc. "Image Vault" device, submitted in 1997. This type of submission focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than comprehensive clinical performance studies usually associated with new technologies or higher-risk devices.
Therefore, many of the typical acceptance criteria and study details requested in your prompt (like specific performance metrics, sample sizes for test and training sets, number and qualifications of experts, MRMC studies, and detailed ground truth establishment) are not present in this document. The device is a PACS Image Archive System/Teleradiology System, which primarily deals with storage and viewing of images, not diagnostic interpretation or algorithm-driven analysis of medical images.
Here's an attempt to answer your questions based solely on the provided text, highlighting what information is available and what is missing.
Acceptance Criteria and Device Performance
The document does not explicitly state "acceptance criteria" in the format of quantitative performance metrics for a specific clinical task. Instead, the device's "performance" is demonstrated through a comparison chart showing its specifications against a predicate device, focusing on functional equivalence.
Table of Acceptance Criteria and Reported Device Performance
| Criterion/Specification | Image Vault (Reported Performance) | Predicate Device (Kodak Cardiac Archive Station CAS 6000) |
|---|---|---|
| Graphical User Interface | Yes | Yes |
| Patient Demographics | Yes | Yes |
| Display Resolution | 1024x768 | 1024x768 |
| Communications | TCP/IP | TCP/IP |
| Image Review (Features) | Flip/Rotate/Pan/Zoom/Sequential | Flip/Rotate/Pan/Zoom/Sequential |
| CD Writers Supported | Yamaha CDR400, Yamaha CDR100 | Kodak PCD Writer 600 |
| CD Labeling | Yes | Yes |
| Temporary Image Storage | Yes | Yes |
| Image Archiving (Hard Disk) | Yes | Yes |
| Image Archiving (Magnetic Tape) | Yes | No |
| Magnetic Tape Drives Supported | ADIC Scalar 218, ADIC Scalar 448 | No |
| Platform | PC | PC |
| DICOM compliance verification | Images compressed are properly identified per DICOM standard. (Implied) | Not explicitly stated, but assumed for equivalence. |
Note: The "acceptance criteria" here are inferred from the comparison chart as the functional specifications and capabilities deemed equivalent to the predicate device. The performance is the Image Vault meeting these specifications.
Study Details
-
Sample size used for the test set and the data provenance:
- Sample Size: Not specified. The document states "Extensive testing of the software package has been performed by programmers, by non-programmers, and by potential customers." This indicates validation testing, but no specific sample size (e.g., number of images, number of patient studies) for a formal test set is provided.
- Data Provenance: Not specified. It's unclear if specific medical image datasets were used for validation, and if so, their origin (country, retrospective/prospective). The device handles "cardiac images," but no specific details about the types or sources of these images for testing are given.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable/Not specified. This device is an archive/display system, not intended for diagnostic interpretation by users. Its functionality relates to storing, retrieving, and displaying images, not generating diagnostic output that would require "ground truth" to be established by medical experts for performance evaluation. The "potential customers" who tested the software are mentioned, but their qualifications and role in establishing ground truth are not detailed.
-
Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- Not applicable/Not specified. Given the nature of the device and the lack of a defined "test set" for diagnostic performance evaluation, an adjudication method for ground truth is not 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. An MRMC study was not conducted. This device is an image archive and display system; it does not incorporate AI or assist human readers in diagnostic tasks. Its purpose is to store and retrieve images, with a simple display utility "not intended to be used for diagnosis."
-
If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- Not applicable. The device is software for PACS archive and display; it does not involve algorithms for standalone diagnostic performance. The validation focused on the software's ability to "send & receive, display, store, and archive DICOM... images." It is stated that "software is only used for control purposes and has no bearing on image quality. There is no image processing used with this software."
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not applicable/Not specified. As the device's function is archiving and display, not diagnostic interpretation, the concept of "ground truth" for clinical accuracy is not relevant in this submission. Validation focused on functional correctness (e.g., proper storage, retrieval, display of images, DICOM compliance). The statement "Images that are compressed are properly identified in the images information as being compressed as specified by the entire DICOM standard" implies that DICOM standard compliance served as a form of "ground truth" for specific functionalities.
-
The sample size for the training set:
- Not applicable/Not specified. This device is not an AI/ML algorithm that requires a "training set" in the conventional sense. It is a software package for data management.
-
How the ground truth for the training set was established:
- Not applicable. No training set for an AI/ML algorithm is described.
§ 892.2020 Medical image communications device.
(a)
Identification. A medical image communications device provides electronic transfer of medical image data between medical devices. It may include a physical communications medium, modems, and interfaces. It may provide simple image review software functionality for medical image processing and manipulation, such as grayscale window and level, zoom and pan, user delineated geometric measurements, compression, or user added image annotations. The device does not perform advanced image processing or complex quantitative functions. This does not include electronic transfer of medical image software functions.(b)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to § 892.9.