(23 days)
The InkSpace Imaging Small Body Array is a receive-only coil, used for obtaining diagnostic images of general human anatomy in adult and pediatric patients such as cardiac, spine, shoulder, elbow, knee, foot, and prostate in Siemens 1.5T magnetic resonance imaging systems. These images, when interpreted by a trained physician, yield information that may assist in diagnosis.
The InkSpace Imaging Small Body Array is a phased array, receive-only coil intended to work with Siemens 1.5T MRI scanners for body MR imaging examinations. The coil is comprised of 24 total channels, with two individual flexible pads, each containing 12 individual elements. The elements are flexible to conform to patients' anatomies.
The provided text does not describe a study involving a device that uses AI or machine learning, nor does it provide a comparative effectiveness study with human readers. Instead, it describes a Magnetic Resonance (MR) Receive-only Coil and its substantial equivalence to a predicate device.
Therefore, many of the requested elements about acceptance criteria for AI performance, sample sizes for test/training sets, expert adjudication, MRMC studies, and ground truth establishment are not applicable based on the provided document.
However, I can extract information related to the device's performance testing from the document:
Device: InkSpace Imaging Small Body Array (a receive-only coil for MRI systems)
1. Table of acceptance criteria and the reported device performance:
The document refers to performance testing in accordance with the "Magnetic Resonance (MR) Receive-only Coil – Performance Criteria for Safety and Performance Based Pathway" guidance document. It lists the tests performed, implying these tests are the acceptance criteria. The performance results are stated to have met established specifications.
| Acceptance Criteria (Nonclinical Test) | Reported Device Performance |
|---|---|
| Image Signal to Noise (SNR) | Met established specifications |
| Image Uniformity | Met established specifications |
| Surface Heating | Met established specifications |
| Decoupling Circuit | Met established specifications |
| EMC - Immunity, Electrostatic Discharge | Met established specifications |
| General Electrical/Mechanical Safety | Met established specifications |
| Acquired Image Quality | Met established specifications (Analyzed based on sample clinical images and reviewed by a board-certified radiologist) |
2. Sample size used for the test set and the data provenance:
- Sample Size: The document does not specify exact sample sizes (e.g., number of coils tested, number of images reviewed) for the nonclinical tests. It refers to "sample clinical images" for the acquired image quality analysis.
- Data Provenance: Not specified in terms of country of origin. The test data are generated from nonclinical bench testing and analysis of sample clinical images. The data is retrospective in the sense that the images are existing data used for evaluation.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Number of Experts: "a board-certified radiologist" (implies one, though it could be a general statement about the type of expert).
- Qualifications: "board-certified radiologist."
4. Adjudication method for the test set:
- The document mentions a "review of their clinical quality by a board-certified radiologist." This implies a single expert review, not an adjudication process involving multiple experts for consensus.
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. An MRMC study was not performed. This device is a passive component (MRI coil), not an AI/ML algorithm that assists human readers.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- N/A. This is not an algorithm; it's hardware. Its performance is assessed through its physical and electromagnetic properties, with the final output (images) being subject to human interpretation.
7. The type of ground truth used:
- For "Acquired Image Quality," the ground truth was established by the subjective "clinical quality" assessment by a "board-certified radiologist." Other tests (SNR, Uniformity, etc.) likely relied on objective measurements against engineering specifications.
8. The sample size for the training set:
- N/A. This device does not involve a training set as it's a hardware component, not an AI/ML model.
9. How the ground truth for the training set was established:
- N/A. (See point 8)
§ 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.