(70 days)
The DICOM Archival Server (DAS)/DICOM IntraNET Service (DIS), software product, provides for the centralized management and storage of radiographic studies, replacing or augmenting the historical film-based processes.
The DICOM Archival Server (DAS) software product provides for the centralized management and storage of radiographic studies, replacing or augmenting the historical filmbased processes. The operating system utilizes Microsoft @ Windows NT @ Release 4.0 and later. This environment provides for all communication (i.e. TCP/IP), file storage (e.g. NTFS), security management, and display management processes. It also insulates the DAS from detail hardware management processes, allowing customers to make whatever choices in these areas best serve them. This environment supports a variety of back-up/restore processes which may be used at the customer's discretion. The Microsoft© SQL Server product is used as the primary management component for the archive contents and for responding to requests. whether they be in the form of SQL Server direct requests or via the DICOM Query/Retrieve process. This software-only product is fully DICOM-3.0 compliant for the following DICOM Service Classes: Verification (SCP), Storage (SCP and SCU), Query/Retrieve (SCP), Print (SCU), Patient Management (SCU), Study Management (SCU), Results Management (SCU). This product consists only of software and executes in a Windows© NT© environment on Intel© platforms. This product, when combined with appropriate server platform hardware and operating system, falls into the FDA proposed designation of a PACS system as Class II (Special Controls). The following attributes are noted: this product accepts, stores and transfers images; no image processing is done by this product; no compressed images are accepted for storage by this product; this product will supply compressed images as requested by client products, at the level of compression requested by those clients; no image viewing is provided by this product; the reliability of the product is dependent upon the hardware on which the software is installed, and is determined by the installation; disaster recovery attributes of an installation are dependent upon installationdetermined policies, and are totally under the control of the installation. The DAS product contains the following advanced features to facilitate rapid movement of images to desired targets within a PACS system: automatic image forwarding and automatic image printing. Automatic image forwarding enables a policy-managed forwarding process wherein each image received by the DAS will automatically be forwarded to one or more target workstations based upon the following image contents: modality station name, body part, referring physician, and modality type. Automatic image printing is an extension of the automatic forwarding process wherein the forwarding target is a DICOM Print Service Class Provider. The DICOM IntraNET Service (DIS) software product provides for centralized distribution of stored radiographic studies. DIS is an integrated client-server software system designed to allow secured access to radiographic images by licensed medical professionals. The software's server and client ends rely on off-the-shelf Windows NT software with compatible Intel hardware. The client end accesses images through a WEB Browser similar to Microsoft's Internet Explorer through a Query/Retrieve process. The DIS system becomes an extended Viewer attached to a secured internal network or an extended secured Intranet using TCP/IP communications protocol. Full resolution, lossless images are always available to the Client Browser. If lossy compression where used, similar to our equivalent products listed in item 10; Kodak, Base-Ten, or Autocyt, we would recommend that they be used for secondary viewing only and not for diagnostic interpretation. This DIS software is intended to provide the means for medical professionals to display data generated by medical scanning devices on a personal computer or workstation. Competent healthcare professionals would reasonably be expected to exercise judgment in use of this information.
Here's an analysis of the provided text regarding the acceptance criteria and study for the Parameter Developments, DICOM Archival Server (DAS)/DICOM IntraNET Service (DIS):
It is important to note that the provided text is a 510(k) summary for a medical device (Picture Archiving and Communications System - PACS) and does not contain information about specific acceptance criteria or a dedicated performance study with a test set, ground truth, or expert readers in the way modern AI/ML device submissions typically do.
This document predates widespread use of AI/ML in medical imaging and focuses on demonstrating substantial equivalence to predicate devices based on functionalities, technical attributes, and safety/hazard analysis for a software-only product primarily concerned with image storage, retrieval, and management.
Therefore, much of the requested information (acceptance criteria in terms of metrics like sensitivity/specificity, sample sizes for test/training sets, expert qualifications, adjudication methods, MRMC studies, standalone performance with metrics, and ground truth types) is not present in this document.
However, I can extract the available relevant information and explain why other requested details are missing.
Analysis of Acceptance Criteria and Study Information
1. Table of Acceptance Criteria and Reported Device Performance
As mentioned, this document does not present explicit, quantitative acceptance criteria that would typically be seen for a diagnostic AI/ML device (e.g., minimum sensitivity, specificity, AUC). Instead, the "acceptance criteria" are implicitly met by demonstrating substantial equivalence to predicate devices through functional comparison and adherence to DICOM standards and safety considerations.
The "reported device performance" is descriptive of its capabilities rather than quantitative performance against a specific clinical task.
| Acceptance Criterion (Implicitly Met) | Reported Device Performance / Justification |
|---|---|
| Functional Equivalence | Performs centralized management and storage of radiographic studies, replacing or augmenting film-based processes. |
| DICOM 3.0 Compliance | Fully DICOM 3.0 compliant for Verification (SCP), Storage (SCP and SCU), Query/Retrieve (SCP), Print (SCU), Patient Management (SCU), Study Management (SCU), and Results Management (SCU) Service Classes. |
| Operating Environment | Operates on Microsoft® Windows NT® Release 4.0 and later, leveraging its communication (TCP/IP), file storage (NTFS), security, and display management. |
| Image Handling | Accepts, stores, and transfers images. No image processing. Does not accept compressed images for storage (only uncompressed). Can supply compressed images as requested by clients. No image viewing provided by this specific product (client-side). |
| Data Integrity (Lossless) | Only uncompressed images accepted for storage. Lossless compression may be used by Windows NT. Lossy compression never used for storage. Can transmit uncompressed, lossless, or lossily compressed images to clients as requested. |
| Safety and Hazard Analysis | Hazard analysis performed throughout development. Concluded "Level of Concern" is "Minor" as failures related to computer system components are not expected to cause patient death or injury. |
| Conformity to Standards | Standard off-the-shelf hardware (UL, CSA approved, FCC rules, DHHS Radiation Performance Standards). |
| Substantial Equivalence | Demonstrated against predicate devices (Fuji FCR DMS, Lockheed ECHONET, Kodak KDS Medical Image, Base Ten UPACS, Autocyt AMICAS) based on hardware, OS, network protocols, storage types, and DICOM functionalities. |
| Intended Use | Provides means for medical professionals to display data generated by medical scanning devices on a personal computer or workstation, with competent healthcare professionals exercising judgment. |
2. Sample size used for the test set and the data provenance
- Not Applicable / Not Provided. This document describes a PACS software system for storage and management, not a diagnostic algorithm that would be tested on a dataset of patient images for a specific clinical task. Therefore, there's no "test set" in the context of clinical performance metrics.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Not Applicable / Not Provided. As there is no test set for clinical performance, there is no need for experts to establish ground truth.
4. Adjudication method for the test set
- Not Applicable / Not Provided. No test set for clinical performance.
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, not done. This device is a PACS system for image archival and distribution, not an AI-assisted diagnostic tool.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- No, not done. This device is a PACS system, not a diagnostic algorithm. Its "performance" is in its functional capabilities (storage, retrieval, DICOM compliance, etc.), which are demonstrated through technical verification and substantial equivalence to predicates, not through standalone clinical performance.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)
- Not Applicable / Not Provided. There's no diagnostic ground truth relevant to the functionality of this PACS system. Its primary "truth" is its adherence to DICOM standards and its ability to store and retrieve images reliably, which would be verified through technical implementation and testing, not clinical "ground truth."
8. The sample size for the training set
- Not Applicable / Not Provided. This is a software system for image management, not a machine learning model that requires a training set.
9. How the ground truth for the training set was established
- Not Applicable / Not Provided. No training set.
Summary of the Document's Approach:
The 510(k) submission for the DICOM Archival Server (DAS)/DICOM IntraNet Service (DIS) primarily relies on:
- Functional Description: Clearly outlining what the software does (storage, retrieval, DICOM services).
- Safety Analysis: Identifying potential hazards (e.g., system component failure) and concluding a "Minor Level of Concern" for patient safety given its role as a PACS.
- DICOM Compliance: Stating full compliance with DICOM 3.0 service classes.
- Substantial Equivalence: Comparing its features and functionalities demonstrably against several legally marketed predicate PACS devices. This is the core "study" or justification for its market entry. The tables in Section 10 ("Substantial Equivalence") serve as the performance comparison in this context.
This approach is typical for medical devices of this vintage and type (PACS as infrastructure, not diagnostic AI), where the focus is on safety, interoperability, and proper functioning as a data management system rather than diagnostic accuracy.
§ 892.2010 Medical image storage device.
(a)
Identification: A medical image storage device is a hardware device that provides electronic storage and retrieval functions for medical images. Examples include electronic hardware devices employing magnetic and optical discs, magnetic tapes, and digital memory.(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.