(43 days)
The 1.5T GE and Siemens DuoFLEX® Coil Suites and the 3.0T GE and Siemens DuoFLEX® Coil Suites are indicated for use on the order of a physician in conjunction with a 1.5T GE Healthcare and Siemens Healthcare Magnetic Resonance Scanner system and a 3.0T GE Healthcare and Siemens Healthcare Magnetic Resonance Scanner system to produce 2D and 3D images that when interpreted by a trained physician vield information that may assist in diagnosis.
The MR Instruments FC 1520G-8R, FC 3000G-8R, FC 1500S-8R, and FC 3000S-8R Coil Suites are identical with the only exceptions:
FC 1520G-8R (GE MR Systems @ 63.86MHz with GE System Connector Plug)
FC 3000G-8R (GE MR Systems @ 127.73MHz with GE System Connector Plug)
FC 1500S-8R (Siemens MR Systems @ 63.6MHz with Siemens System Connector Plug)
FC 3000S-8R (Siemens MR Systems @ 123.23MHz with Siemens System Connector Plug)
The coil suites include two different sets of coil paddles and a single, shared preamplifier box (the "Connector Box") with a system connector. One set contains a pair of coil paddles containing four 24cm loops/channels per paddle for a total of eight channels for the two paddles. The second set contains a pair of coil paddles containing four 10cm loops/channels per paddle for a total of eight channels for the two paddles. These paddles connect to the same connector box containing eight preamplifiers; only one pair can be connected at a time. The coil design for all the paddles has the same, simple antennae design and the same system connectivity configuration. The coils can be used in the following configurations:
24cm by itself
24cm with 24cm
24cm with 10cm
10cm with 10cm
The aforementioned configurations provide 4 or 8 channels for imaging.
The provided document is a 510(k) premarket notification for the "DuoFLEX® Coil Suite," which is an MR diagnostic device (specialty magnetic resonance coil). This document does not describe an AI/ML-driven device or an acceptance study typical for such devices. Instead, it details the substantial equivalence of a new version of an MR coil to a previously cleared predicate device.
Therefore, many of the requested criteria related to AI/ML model acceptance studies (like sample size for training/test sets, expert adjudication, MRMC studies, ground truth establishment for training, etc.) are not applicable to this document. The "study" described is a set of non-clinical bench tests and performance testing with phantoms and volunteer scans to demonstrate substantial equivalence, not a clinical study to prove diagnostic performance improvement by an AI.
Here's an analysis of what is available in the document, mapped against the closest applicable aspects of your request, and explicitly stating what is not present:
Device: DuoFLEX® Coil Suite (Magnetic Resonance Imaging Coil)
Type of Study Conducted: Non-clinical bench testing, performance testing with phantoms, volunteer scans, and comparison to a predicate device to demonstrate substantial equivalence. This is not an AI/ML model performance study.
1. Table of Acceptance Criteria and Reported Device Performance:
The document describes the tests performed to demonstrate substantial equivalence rather than explicit acceptance criteria with numerical performance targets in a table format. The "acceptance" is based on the new device performing equivalently to the predicate and meeting safety standards.
| Acceptance Criteria (Implied from the "Non-Clinical Testing" section) | Reported Device Performance / Outcome |
|---|---|
| EMC and electrical safety testing | Tests conducted; presumed to have met standards for safety. |
| System safety testing | Tests conducted; presumed to have met standards for safety. |
| Performance testing with phantoms | Tests conducted; outcomes used to support substantial equivalence. |
| Predicate device comparison tests | Tests conducted; results support substantial equivalence. |
| Volunteer scans | Scans conducted; used to support substantial equivalence. |
| IEC 60601-1 Compliance (various tests) | Tests conducted; presumed to have met standards for medical electrical equipment safety. |
| Biocompatibility for patient contact materials | Tests conducted; presumed to have met standards. |
| NEMA MS-1 Signal to Noise Ratio (SNR) | Test conducted; used for performance evaluation against predicate. |
| Image Uniformity Comparison | Test conducted; used for performance evaluation against predicate. |
| Evaluation of Sample Clinical Images (Clinical Evaluation Testing) | Test conducted; used to support substantial equivalence. |
| Design Validation (in actual and simulated use settings) | Performed; results support substantial equivalence and safety. |
2. Sample size used for the test set and the data provenance:
- Test Set (for performance evaluation):
- Phantoms: Not specified, but phantom testing was performed.
- Volunteer Scans: "Volunteer scans" were conducted. The number of volunteers is not specified.
- Clinical Images: "Sample Clinical Images" were evaluated. The number of images/patients is not specified.
- Data Provenance: Not explicitly stated for volunteers or clinical images, but it's implied these were internal studies conducted by MR Instruments, Inc. No mention of country of origin or whether prospective/retrospective. Given it's a device clearance, the data would typically be gathered for the purpose of the submission.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable in the context of an AI/ML ground truth study. The evaluation involved "Evaluation of Sample Clinical Images (Clinical Evaluation Testing)" and comparison to the predicate. The document doesn't detail how "ground truth" (in the sense of a diagnostic label) was established, as the device is a coil for producing images, not interpreting them. The assessment would likely be on image quality, signal consistency, safety, etc., compared to the predicate, rather than diagnostic accuracy.
4. Adjudication method for the test set:
- Not applicable. No mention of expert adjudication as this is not a diagnostic AI performance study.
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 MRMC study was done. This is not an AI-assisted diagnostic device. The study focused on demonstrating substantial equivalence of the new MR coil to an existing predicate coil. The document explicitly states: "This technology is not new; therefore, a clinical study was not considered necessary prior to release. Additionally, there was no clinical testing required to support the medical devices as the indications for use is equivalent to the predicate device. The substantial equivalence of the devices is supported by the non-clinical testing."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Partially applicable, but not for an AI algorithm. The "standalone" performance here refers to the performance of the MR coil itself (e.g., SNR, image uniformity, safety) as measured through bench tests, phantoms, and volunteer scans, independent of human interpretation for diagnostic purposes. These non-clinical tests demonstrate the coil's intrinsic characteristics.
7. The type of ground truth used:
- Not explicitly defined as "ground truth" in the diagnostic sense. For this device (an MR coil), "ground truth" implicitly relates to physical properties (e.g., magnetic field homogeneity, signal generation), safety parameters, and image quality characteristics (SNR, uniformity) as measured by calibrated equipment and compared against known standards or the predicate device's performance. The "volunteer scans" and "sample clinical images" would be assessed for image quality and usability for diagnostic purposes, not for confirming a specific diagnosis.
8. The sample size for the training set:
- Not applicable. This document describes an MR coil, not an AI/ML model that requires a training set.
9. How the ground truth for the training set was established:
- Not applicable. As above, this is not an AI/ML model.
§ 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.