(162 days)
uMR Jupiter is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces sagittal, transverse, coronal, and oblique cross sectional images, and spectroscopic images, and that display internal anatomical structure and/or function of the head, body and extremities. These images and the physical parameters derived from the images when interpreted by a trained physician yield information that may assist the diagnosis. Contrast agents may be used depending on the region of interest of the scan.
The device is intended for patients > 20 kg/44 lbs.
uMR Jupiter is a 5T superconducting magnetic resonance diagnostic device with a 60cm size patient bore and 8 channel RF transmit system. It consists of components such as magnet, RF power amplifier, RF coils, gradient power amplifier, gradient coils, patient table, spectrometer, computer, equipment cabinets, power distribution system, internal communication system, and vital signal module etc. uMR Jupiter is designed to conform to NEMA and DICOM standards.
Here's a breakdown of the acceptance criteria and the study proving the device meets them, based on the provided FDA 510(k) submission information for the uMR Jupiter.
Acceptance Criteria and Reported Device Performance
The acceptance criteria for the uMR Jupiter largely revolve around its non-inferiority to the predicate device (uMR Omega) and its ability to produce diagnostic quality images while ensuring safety. The performance evaluation focused on various aspects of MRI image quality and the functionality of new or enhanced features, especially the AI-assisted Compressed Sensing (ACS).
| Acceptance Criteria Category | Specific Criteria/Tests | Reported Device Performance (uMR Jupiter) |
|---|---|---|
| General Comparison to Predicate | The proposed device should have similar indications for use, performance, safety equivalence, and effectiveness as the predicate device. Differences should not raise new safety and effectiveness concerns. | The submission concludes that "the proposed device has similar indications for use, performance, safety equivalence, and effectiveness as the predicate device. The differences above between the proposed device and predicate device do not affect the intended use, technology characteristics, safety, and effectiveness. And no issues are raised regarding to safety and effectiveness." This broadly states that all differences (field strength, bore dimensions, magnet homogeneity, gradient amplitude, RF system, coils, etc.) were evaluated and deemed not to raise new safety or effectiveness concerns. |
| Image Quality (Non-Clinical) | Conformance to NEMA standards (MS 1, MS 2, MS 3, MS 5, MS 6, MS 9) for SNR, geometric distortion, image uniformity, slice thickness, and characterization of phased array coils. | Non-clinical testing, including image performance tests, were conducted to verify that the proposed device met all design specifications. This implies adherence to the mentioned NEMA standards. |
| Safety (Non-Clinical) | Conformance to IEC 60601-1 (General), IEC 60601-1-2 (EMC), IEC 60601-2-33 (Magnetic Resonance Equipment), IEC 60825-1 (Laser Safety), IEC 60601-1-6 (Usability), IEC 62304 (Software Life Cycle), IEC 62464-1 (Image Quality Parameters), NEMA MS 8 (SAR), NEMA MS 10 (Local SAR), NEMA MS 14 (RF Coil Heating), IEC 60601-4-2 (EMC Immunity). Control of peripheral nerve stimulation (PNS) and cardiac stimulation. SAR control for patients > 20kg. | Electrical safety and EMC tests were performed, claiming conformance to the listed IEC and NEMA standards. A volunteer study was conducted to determine nerve stimulation thresholds, and observed parameters were used to set PNS threshold levels as required by IEC 60601-2-33. The device's software controls SAR based on simulations for humans at least 20kg. |
| Software Functionality | Functionality of new/enhanced features (e.g., Inline T2 Mapping using MASS, CASS, PASS, MoCap-Monitoring). | Performance evaluation reports were provided for ACS, 3D ASL, MoCap-Monitoring, FACT, 2D Flow, CEST, T1rho, Multiband, Inline T1 mapping, Inline T2 mapping, Inline T2* mapping, Liver MRS, Prostate MRS, and Brain MRS. The submission explicitly states that MASS, CASS, and PASS are "substantially equivalent" to existing techniques and MoCap-Monitoring provides real-time motion monitoring. |
| AI (ACS) Performance | ACS (AI-assisted Compressed Sensing) should perform at least equivalently to Compressed Sensing (CS) in terms of SNR and resolution. Image qualities (contrast, uniformity) should be maintained compared to fully sampled data (golden standard). Structural measurements on paired images (ACS vs. fully sampled) should be significantly the same. Performance should be equivalent to ACS on the predicate device (uMR Omega). | ACS on uMR Jupiter was shown to "perform better than CS by measuring SNR and resolution" across diverse demographics and pathological variations. Results demonstrated that ACS "maintained image qualities, such as contrast and uniformity, as compared against fully sampled data as golden standards." Structural measurements verified that ACS and fully sampled images of the same structures were "significantly the same." The test results demonstrate that "ACS on uMR Jupiter performs equivalently to that on uMR Omega." |
| Clinical Image Quality | The device should generate diagnostic quality images in accordance with MR guidance on premarket notification submissions. | Sample clinical images for all clinical sequences and coils were reviewed by three U.S. board-certified radiologists, and it was shown that the proposed device can generate diagnostic quality images comparable to the predicate. |
| Biocompatibility | Conformance to ISO 10993-5 (In vitro cytotoxicity), ISO 10993-10 (Skin Sensitization), ISO 10993-23 (Irritation), and ISO 10993-1 (Evaluation and testing within a risk management process). | Claims conformance to the listed ISO 10993 standards. |
| Risk Management | Conformance to ISO 14971 (Application of risk management to medical devices). | Claims conformance to ISO 14971. |
| Quality System | Conformance to 21 CFR Part 820 (Quality System Regulation). | Claims conformance to 21 CFR Part 820. |
Study Details (Focusing on ACS as the AI component)
The document primarily details the performance evaluation of the AI-assisted Compressed Sensing (ACS) module, as it represents a key software enhancement.
1. A table of acceptance criteria and the reported device performance (see table above).
2. Sample sizes used for the test set and the data provenance:
- Test Set (for ACS performance): 25 subjects.
- Demographic Breakdown:
- Gender: 15 Male, 10 Female
- Age: 5 (18-28), 7 (29-40), 13 (>41)
- Ethnicity: 4 White, 21 Asian
- BMI: 2 Underweight (24.9)
- Data Provenance: "coming from different countries with diverse demographic distributions covering various genders, age groups, ethnicities, and BMI groups." The document does not explicitly state the specific countries, but the ethnicity breakdown suggests a diverse geographic origin. It's implied this was prospectively collected as a designated test set, and crucially, it was collected independently from the training dataset, with separated subjects and during different time periods.
- Demographic Breakdown:
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- For the ACS performance evaluation: The "ground-truth" for ACS was derived from fully-sampled k-space data, which was then transformed into image space. This indicates a technical, objective ground truth based on the full data acquisition, not directly on expert consensus for the ACS performance metrics (SNR, resolution, contrast, uniformity).
- For overall clinical image quality review: "Sample clinical images for all clinical sequences and coils were reviewed by three U.S. board-certified radiologists comparing the proposed device and predicate device." No specific years of experience are listed beyond "board-certified." This review established diagnostic quality.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- For ACS performance metrics (SNR, resolution, contrast, uniformity): No explicit adjudication method among experts is mentioned for these quantitative metrics, as the ground truth was "fully sampled data" which is a direct technical reference.
- For the overall clinical image quality review by radiologists: While three radiologists reviewed images, the document states "it was shown that the proposed device can generate diagnostic quality images," implying a consensus or satisfactory individual assessment, but no specific adjudication rule (e.g., majority vote, binding senior reviewer) is detailed.
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 Multi-Reader Multi-Case (MRMC) comparative effectiveness study was explicitly described in the provided text for human readers assisted by AI. The evaluation focused on the standalone performance of the ACS algorithm itself relative to conventional CS and fully sampled data, and the overall clinical image quality review. There is no information about an effect size related to human reader improvement with AI assistance.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Yes, a standalone performance evaluation of the ACS algorithm was a primary component. The comparison of ACS against conventional CS and fully sampled data (golden standard) in terms of SNR, resolution, contrast, uniformity, and structural measurements directly assesses the algorithm's performance without a human in the loop for those specific quantitative metrics. The phrase "ACS on uMR Jupiter was shown to perform better than CS by measuring SNR and resolution" confirms this.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Technical/Objective Ground Truth: For the ACS performance evaluation, the ground truth was fully-sampled k-space data that was converted to image space. This serves as the "golden standard" against which the accelerated ACS images were compared for quantitative metrics like SNR, resolution, contrast, and uniformity.
- Expert Consensus (Implicit/Qualitative): For the overall assessment of diagnostic image quality, the review by three U.S. board-certified radiologists served as the qualitative ground truth for "diagnostic quality."
8. The sample size for the training set:
- Training Dataset (for AI module in ACS): Collected from 35 volunteers, comprising 24 males and 11 females, aged 18 to 60.
9. How the ground truth for the training set was established:
- The ground truth for the AI module's training data for ACS was established using fully-sampled k-space data converted to image space.
- "Fully-sampled k-space data were collected and transformed to image space as the ground-truth."
- "Input data [for training] were generated by sub-sampling the fully-sampled k-space data with different parallel imaging acceleration factors and partial Fourier factors."
- "All data were manually quality controlled before included for training." This manual quality control step likely involved expert review to ensure the quality of the "fully sampled" ground truth images.
§ 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.