(177 days)
Diffusion Weighted EPI imaging produces magnetic resonance (MR) images whose contrast is dependent on the local diffusion coefficient of water. Diffusion Weighted EPI can be useful in visualizing the apparent loss of diffusion (mobility) by water molecules in brain tissues affected by acute stroke. Diffusion Weighted EPI is more accurate than conventional MRI pulse sequence (ie Fast FLAIR and Fast Spin Echo) in identifying the occurrence of acute stroke.
The Diffusion Weighted EPI Imaging Option provides an additional imaging option to the Echo Planar Imaging pulse sequence. The DW-EPI option is a single shot EPI pulse designed to create images that differentiates tissues with restricted diffusion from tissues with normal diffusion.
Here's a summary of the acceptance criteria and study information for the GE Medical Systems Diffusion Weighted EPI Imaging Option based on the provided text:
Important Note: The provided document is a 510(k) summary for a medical device submitted in 1997. At that time, the regulatory requirements for showing performance and clinical benefit were significantly different (and generally less stringent) than today's standards for AI/ML devices. Therefore, much of the information typically expected for a modern AI device's acceptance criteria and study data (e.g., detailed performance metrics like sensitivity/specificity, specific ground truth methods for clinical endpoints, robust training/test set details, MRMC studies) is not present in this document. The focus of this submission is primarily on demonstrating substantial equivalence to a predicate device and basic safety.
Acceptance Criteria and Device Performance
| Acceptance Criteria | Reported Device Performance |
|---|---|
| Safety: Device meets international medical equipment safety standards for Magnetic Resonance Systems. | Evaluation testing confirmed compliance with IEC 601-2-33 International medical equipment safety standard for Magnetic Resonance Systems. The opinion of GE is that the Diffusion Weighted EPI Imaging Option does not result in any new potential hazards. |
| Accuracy: Differentiation of tissues with restricted diffusion from tissues with normal diffusion. | The device is designed to create images that differentiate tissues with restricted diffusion from tissues with normal diffusion. |
| Clinical Utility: Useful in visualizing apparent loss of diffusion (mobility) of water molecules in brain tissues affected by acute stroke. | The device produces MR images whose contrast is dependent on the local diffusion coefficient of water. It "can be useful" in visualizing the apparent loss of diffusion (mobility) by water molecules in brain tissues affected by acute stroke. The document states, "Diffusion Weighted EPI is more accurate than conventional MRI pulse sequence (ie Fast FLAIR and Fast Spin Echo) in identifying the occurrence of acute stroke." (However, the study details to support this comparative accuracy claim explicitly are missing/not detailed in the summary.) |
| Substantial Equivalence: To predicate device. | The Diffusion Weighted EPI Imaging Option is substantially equivalent to the currently marketed Siemens Medical System Diffusion Weighted MR Imaging Option (510k #K971055). This implies the predicate already met accepted safety and performance levels deemed adequate by the FDA. |
Study Details
-
Sample sizes used for the test set and the data provenance:
- Test Set Sample Size: Not explicitly stated. The "Summary of Studies" section only mentions "Evaluation testing confirmed accuracy statements in the User Manual." It does not provide details on patient cohorts, case numbers, or specific performance metrics from a test set.
- Data Provenance: Not specified.
-
Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not specified. The document does not detail how "accuracy statements" were confirmed, nor does it refer to specific clinical ground truth established by experts for performance evaluation beyond general statements about its utility in identifying acute stroke. The primary "ground truth" for regulatory approval appears to be equivalence to the predicate device.
-
Adjudication method for the test set:
- Not specified.
-
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 MRMC comparative effectiveness study is reported. The device is an "imaging option" (a pulse sequence), not an AI-based interpretation tool designed to assist human readers, in the modern sense. The claim is that the imaging sequence itself is "more accurate than conventional MRI pulse sequence" in identifying acute stroke, not that it improves human reader performance through AI assistance.
-
If a standalone (i.e., algorithm only without human-in-the-loop performance) was done:
- This is not an "algorithm-only" or AI device in the contemporary sense. It's an MR imaging pulse sequence. Its "performance" is inherent in the image contrast and information it provides. The study primarily focuses on safety and functional performance of the imaging sequence itself, not on an algorithm's diagnostic output.
-
The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not explicitly stated for clinical accuracy. The document mentions "accuracy statements in the User Manual" were confirmed, but the basis for these statements (e.g., against pathology, clinical follow-up, or expert consensus) is not detailed. For the regulatory submission, the primary "ground truth" was the predicate device and its established safety and performance.
-
The sample size for the training set:
- Not applicable as this is an MR imaging pulse sequence, not an AI/ML algorithm that requires a training set in the current understanding. Its development likely involved engineering, physics, and empirical testing for image quality and contrast, rather than data-driven machine learning.
-
How the ground truth for the training set was established:
- Not applicable for the same reason as #7.
§ 892.1000 Magnetic resonance diagnostic device.
(a)
Identification. A magnetic resonance diagnostic device is intended for general diagnostic use to present images which reflect the spatial distribution and/or magnetic resonance spectra which reflect frequency and distribution of nuclei exhibiting nuclear magnetic resonance. Other physical parameters derived from the images and/or spectra may also be produced. The device includes hydrogen-1 (proton) imaging, sodium-23 imaging, hydrogen-1 spectroscopy, phosphorus-31 spectroscopy, and chemical shift imaging (preserving simultaneous frequency and spatial information).(b)
Classification. Class II (special controls). A magnetic resonance imaging disposable kit intended for use with a magnetic resonance diagnostic device only is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 892.9.