(14 days)
I4 (Integrated Intelligent Imaging Informatics) is an image management system intended to be used by trained professionals, including but not limited to radiologists.
I4 (Integrated Intelligent Imaging Informatics) system is a software package used with general purpose computing hardware to acquire, store, distribute, process and associated data throughout a clinical environment. The software performs digital image processing, measurement, manipulation and quantification of images, communication and storage.
This device is not to be used for mammography.
I4 (Integrated Intelligent Imaging Informatics) is an image management system intended to be used by trained professionals, including but not limited to radiologists.
I4 (Integrated Intelligent Imaging Informatics) system is a software package used with general purpose computing hardware to acquire, store, distribute, process and display images and associated data throughout a clinical environment. The software performs digital image processing, measurement, manipulation and quantification of images, communication and storage.
This device is not to be used for mammography.
I4 (Integrated Intelligent Imaging Informatics) is a medical software system offering a primary interpretation solution for visualization and evaluation a variety of medical images deriving from various imaging modalities as well as non-imaging information. I4 interconnects with clinical imaging and non-imaging data sources to present in addition to images non-imaging data in patient context.
This document (K160315) describes the Philips I4 (Integrated Intelligent Imaging Informatics) system, an image management software package. It focuses on demonstrating substantial equivalence to a predicate device, IntelliSpace PACS 4.x (K111804), rather than proving the device meets specific clinical acceptance criteria for a novel diagnostic task.
Therefore, the requested information regarding "acceptance criteria and the study that proves the device meets the acceptance criteria" in terms of clinical performance metrics, sample sizes for test/training sets, expert ground truth establishment, or MRMC studies for improved reader performance is not present in this 510(k) summary.
The document asserts that the device "Meets the acceptance criteria and is adequate for its intended use" based on non-clinical performance testing and compliance with relevant standards. The "acceptance criteria" here refer to internal development and regulatory compliance criteria, not the performance metrics of a diagnostic AI as one might typically expect with such a request.
Here's a breakdown of what is available based on the provided text, and where information is explicitly missing:
Acceptance Criteria and Reported Device Performance
Acceptance Criteria (as implied by the document):
The device is intended to:
- Function as an image management system for trained professionals (e.g., radiologists).
- Acquire, store, distribute, process, and display images and associated data in a clinical environment.
- Perform digital image processing, measurement, manipulation, and quantification of images, communication, and storage.
- Comply with specified international and FDA-recognized consensus standards (ISO 14971, IEC 62304, IEC 62366-1, NEMA-PS 3.1-PS 3.20 DICOM).
- Follow the FDA guidance "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices."
- Maintain safety and effectiveness comparable to the primary predicate device (IntelliSpace PACS 4.x, K111804).
- Not be used for mammography.
- Incorporate advanced visualization and evaluation capabilities from reference predicate devices.
Reported Device Performance (based on non-clinical testing):
The document states that "I4 (Integrated Intelligent Imaging Informatics) system was tested in accordance with Philips verification and validation processes. Verification and Validation tests have been performed to address intended use, the technological characteristics claims, requirement specifications and the risk management results."
The results of these tests demonstrate that the system:
- "complies with the aforementioned international and FDA-recognized consensus standards and FDA guidance document."
- "Meets the acceptance criteria and is adequate for its intended use."
Detailed Performance Table:
A table of specific quantitative acceptance criteria (e.g., sensitivity, specificity, accuracy, processing speed, image quality metrics) and matching numerical performance results is not provided in this document. The "acceptance criteria" discussed are overarching regulatory and functional compliance rather than specific diagnostic accuracy metrics.
Additional Requested Information:
-
Sample size used for the test set and the data provenance:
- Not explicitly stated for clinical performance. The document refers to "non-clinical performance testing" which would involve software verification and validation, possibly using test data sets, but details on the size or provenance of such data (e.g., if it involved medical images) are not provided. The study is not a clinical study directly evaluating diagnostic accuracy on a patient cohort.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable. As this was a non-clinical performance evaluation focused on substantial equivalence through functional and technological comparison, there was no clinical test set requiring expert ground truth for diagnostic accuracy.
-
Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not applicable. See point 2.
-
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 done. The document explicitly states: "The subject of this premarket submission, I4 (Integrated Intelligent Imaging Informatics) system did not require clinical studies to support equivalence." This indicates no clinical performance study involving human readers or AI assistance in a diagnostic context was conducted or submitted.
-
If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not applicable in the sense of a diagnostic AI algorithm. The I4 system is an image management system (PACS), not a standalone diagnostic AI algorithm. Its performance is evaluated on its ability to acquire, store, process, and display images, and its substantial equivalence to an existing PACS system.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not applicable in a clinical diagnostic sense. The "ground truth" for the non-clinical testing would relate to functional specifications and expected software behavior (e.g., "does the image display correctly?", "does the system store the image as specified?").
-
The sample size for the training set:
- Not applicable. This is a PACS system, not a machine learning model, so there is no training set in the AI sense.
-
How the ground truth for the training set was established:
- Not applicable. See point 7.
In summary, the provided document focuses on demonstrating the substantial equivalence of the I4 system to predicate devices based on functional, technological, and regulatory compliance, rather than clinical performance data from a prospective study evaluating diagnostic accuracy. The "acceptance criteria" referred to are related to internal development and regulatory standards for a PACS system, not clinical diagnostic metrics.
§ 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).