(50 days)
Horizon Medical Imaging is a medical image and information management system that is intended to receive, transmit, store, retrieve, display, print and process digital medical images, digital medical video, and associated medical information from various medical imaging systems.
The medical modalities of these medical imaging systems include, but are not limited to, all modalities supported by ACR/NEMA DICOM 3.0 (specifically including mammographic images).
Horizon Medical Imaging is intended to connect to a variety of storage systems and printers via DICOM and other computer industry standard interfaces and protocols.
Lossy compressed mammographic images and digitized film screen images must not be reviewed for primary image interpretations. Mammographic images may only be interpreted using an FDA approved monitor that offers at least 5 megapixel resolution and meets other technical specifications reviewed and approved by FDA. Horizon Medical Imaging will simply perform normal image manipulations for grayscale and image contrast on mammographic images and will not perform image processing on mammographic images.
Horizon Medical Imaging is indicated for use by trained medical professionals including, but not limited to, radiologists, physicians, and medical technologists. Horizon Medical Imaging is also indicated for use in soft-copy diagnostic interpretation of medical images and video by physicians trained in such practice (specifically including soft-copy diagnostic interpretation of mammographic images).
Horizon Medical Imaging is software which when installed and run on Microsoft Windows 2000 and XP operating systems on commercially available IBM PC compatible computers, hardware components and peripherals, forms a medical image and information management system that receives, transmits, stores, retrieves, displays, prints and processes digital medical images, digital medical video, and associated medical information from various medical imaging systems.
Its core components are:
- High resolution color and grayscale workstations for primary diagnostic interpretation and secondary review of the medical images, video and related information.
- Standard workstations for performing administrative functions of the system.
- Servers for short-term and long-term storage of system data.
- Servers for managing the distribution of system data.
- Components for providing the communication channels between the system's core components and the various medical imaging systems.
- Components for producing hardcopy of medical images and related information.
The provided text is a K043146 510(k) premarket notification for Horizon Medical Imaging, largely focusing on its regulatory approval, intended use, and substantial equivalence to previously cleared PACS systems. It does not contain information about acceptance criteria or a specific study proving the device meets acceptance criteria.
The document states:
- Device Name: Horizon Medical Imaging
- Function: Software for receiving, transmitting, storing, retrieving, displaying, printing, and processing digital medical images, video, and related information.
- Intended Use: For use by trained medical professionals (radiologists, physicians, medical technologists) for soft-copy diagnostic interpretation of medical images and video, including mammographic images.
- Classification Name: Picture Archiving and Communications System (PACS) per 21 CFR 892.2050.
- Basis for Clearance: Substantial equivalence to predicate PACS devices (K925965, K973959, K023557).
Therefore, I cannot provide the requested information as it is not present in the given text. The 510(k) summary focuses on demonstrating that the new device is as safe and effective as a legally marketed predicate device, rather than detailing a study with specific acceptance criteria and performance metrics for the new device itself.
A 510(k) submission typically relies on comparison to predicate devices, and only if new technology or intended uses are introduced that significantly differ from predicates would more extensive performance data be required. In this case, the main addition noted is the "intended use of soft-copy official diagnostic interpretation of mammographic images," which is claimed to be equivalent to the GE Medical Systems Centricity PACS Plus.
To answer your specific questions, had the information been present, this is generally how it would be structured:
- Table of acceptance criteria and reported device performance: This would list metrics like image fidelity, measurement accuracy, display accuracy, network throughput, etc., and the corresponding pass/fail criteria and measured values from tests.
- Sample size and data provenance for the test set: Details on how many images/studies were used for performance testing, where they came from (e.g., anonymized patient data from specific hospitals), and if they were collected retrospectively or prospectively.
- Number of experts and qualifications for ground truth: Information on how many specialists reviewed the test data to establish the correct diagnoses or measurements, and their specific expertise (e.g., 3 board-certified radiologists with 15+ years of experience in mammography).
- Adjudication method for the test set: How disagreements among experts were resolved (e.g., a third expert review, majority vote).
- Multi-Reader Multi-Case (MRMC) comparative effectiveness study: If such a study was performed, it would detail the experimental setup (human readers with and without AI assistance), the metrics used (e.g., AUC, sensitivity, specificity), and the quantitative improvement observed by human readers with AI.
- Standalone performance study: If an algorithm-only (AI without human-in-the-loop) performance was evaluated, this section would provide metrics like sensitivity, specificity, F1-score, or AUC.
- Type of ground truth: What the "true" answer was based on (e.g., surgical pathology results, long-term patient outcomes, consensus of highly experienced experts).
- Sample size for the training set: The number of images/cases used to train any AI or machine learning components within the device.
- Ground truth for the training set: How the correct labels or annotations were applied to the training data (e.g., expert annotation, automated labeling, pathology reports).
Since the provided document is a 510(k) summary focused on substantial equivalence to predicate devices, it primarily discusses the device's function, intended use, and comparison to existing products, rather than detailed performance study results against specific acceptance criteria.
§ 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).