(90 days)
The MR Cardiac Analysis Package is intended for use in processing cardiac or cardiovascular MR images, resulting in quantitative and/or qualitative data that can be used to estimate cardiac or cardiovascular function.
The MR Cardiac Analysis Package is indicated for use in association with cardiac and cardiovascular examinations where it is useful or necessary to perform qualitative or quantitative analysis.
The MR Cardiac Analysis Package provides off-line viewing and processing functions for cardiac and cardio-vascular examinations, acquired with Magnetic Resonance, to simplify the process. Historically the user had to outline the end-systolic and end-diastolic ventricle areas in one or all slices. To reduce the time required for the analysis, the MR Cardiac Analysis Package provides a facility for computer assisted detection of the heart boundaries of the left ventricle. The user may edit the computer generated boundary and must always approve the selection. The right ventricle boundary must still be detected manually. To use the MR Cardiac Analysis Package, MR images of the heart are acquired with dedicated MR sequences at multiple slices and different phases of the heart cycle and transferred to the Easyvision Workstation. Processing these images provides data that can be used to estimate various quantitative parameters, e.g., ejection fraction, stroke volume, diastolic and systolic volume, ventricular mass, and wall thickness. It can also provide a display of cardiac functional parameters and permits correction for errors caused by papillary muscle. All quantitative results can be printed in a dedicated, user definable, print layout for cardiac analysis, where analysis data (statistical and graphical) can be reported in combination with the image data. The package will also provide viewing capabilities such as direct access to images, defined by slice location or heart phase, and synchronized movies for the different heart phases and/or slices.
The provided text (K972103) describes the Philips MR Cardiac Analysis Package, an image processing system for cardiac and cardiovascular MR images. However, it does not include a detailed study or specific acceptance criteria with reported device performance metrics. Instead, it focuses on the device's intended use, system description, and regulatory clearance (510(k)).
Therefore, I cannot provide the requested information in a table format with specific performance metrics or detailed study characteristics because that information is not present in the provided document.
Here's what I can infer from the document regarding the study aspects, assuming a study would have been conducted to support substantial equivalence, even if the details aren't explicitly given in this summary:
-
A table of acceptance criteria and the reported device performance:
- Not provided. The document does not specify quantitative acceptance criteria (e.g., accuracy, precision, F1-score) or present a table of reported device performance metrics against such criteria. The 510(k) summary focuses on demonstrating substantial equivalence to predicate devices, which often relies on functional similarity and intended use rather than head-to-head performance comparisons with specific numerical outcome targets.
-
Sample size used for the test set and the data provenance, number of experts for ground truth and their qualifications, and adjudication method:
- Not provided. The document does not mention any specific test set size, data provenance, number or qualifications of experts, or adjudication methods for establishing ground truth for any performance evaluation. It describes the output of the device (quantitative data like ejection fraction, stroke volume, etc.) and highlights its computer-assisted boundary detection feature, but not how the accuracy of these outputs was tested or validated against
a defined ground truth.
- Not provided. The document does not mention any specific test set size, data provenance, number or qualifications of experts, or adjudication methods for establishing ground truth for any performance evaluation. It describes the output of the device (quantitative data like ejection fraction, stroke volume, etc.) and highlights its computer-assisted boundary detection feature, but not how the accuracy of these outputs was tested or validated against
-
If a multi-reader multi-case (MRMC) comparative effectiveness study was done, and the effect size:
- Not provided. There is no mention of an MRMC study or any comparison of human reader performance with or without AI assistance. The device is described as "computer-assisted," where the user can edit and must approve the computer-generated boundary, suggesting a human-in-the-loop approach, but no formal study of its impact on human reader effectiveness is described.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- Partially addressed indirectly. The system provides "computer assisted detection of the heart boundaries of the left ventricle." However, it explicitly states, "The user may edit the computer generated boundary and must always approve the selection." This indicates that a pure standalone (algorithm only) performance, without human intervention, is not the intended or described mode of operation. Therefore, a standalone performance study as typically understood (algorithm output without any human correction) would not be directly relevant to its described use, and no such study is mentioned.
-
The type of ground truth used:
- Not provided. The document does not specify how ground truth would be established for quantitative parameters like ejection fraction, stroke volume, etc., in the context of any validation study. Given the device's output, potential ground truth could involve expert manual measurements, but this is not stated.
-
The sample size for the training set:
- Not provided. The document does not mention any training set or its size. While the device utilizes "computer assisted detection," it does not explicitly state that it employs machine learning or a training phase in the modern sense.
-
How the ground truth for the training set was established:
- Not applicable / Not provided. As no training set or machine learning approach is explicitly described, the method for establishing its ground truth is also not mentioned.
§ 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).