Search Results

The MAGNETOM system is indicated for use as a magnetic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, and that displays the internal structure and/or function of the head or extremities. Other physical parameters derived from the images may also be produced. Additionally, the MAGNETOM system is intended to produce Sodium images for the head and Phosphorus spectroscopic images and/or spectra for whole body, excluding the head. These images and/or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

MAGNETOM Terra and MAGNETOM Terra.X with software syngo MR XA60A include new and modified hardware and software compared to the predicate device, MAGNETOM Terra with software syngo MR E12U. A high level summary of the new and modified hardware and software is provided below: Hardware: New Hardware (Combiner (pTx to sTx), MC-PALI, GSSU control unit, 8Tx32Rx Head coil), Modified Hardware (Main components such as: Upgrade of GPA, New Host computer hardware, New MaRS computer hardware, Upgrade the SEP, The new shim cabinet ASC5 replaces two ACS4 shim cabinets; Other components such as: RFPA, Use of a common MR component which provides basic functionality that is required for all MAGNETOM system types, The multi-nuclear (MNO) option has been modified, OPS module, Cover with UI update on PDD). Software: New Features and Applications (Static B1 shimming, TrueForm (1ch compatibility mode), Deep Resolve Boost, Deep Resolve Gain, Deep Resolve Sharp, Bias field correction (marketing name: Deep RxE), The new BEAT pulse sequence type, BLADE diffusion, The PETRA pulse sequence type, TSE DIXON, The Compressed Sensing (CS) functionality is now available for the SPACE pulse sequence type, The Compressed Sensing (CS) functionality is now available for the TFL pulse sequence type, IDEA, The Scientific Suite), Modified Features and Applications (EP2D DIFF and TSE with SliceAdjust, The Turbo Flash (TFL)), Modified Software / Platform (Stimulation monitoring, "dynamic research labeling"), Other Modifications and / or Minor Changes (Intended use, SAR Calculation and Weight limit reduction for 31P/1H TxRx Flex Loop Coil, X-upgrade for MAGNETOM Terra to MAGNETOM Terra.X, Provide secure MR scanner setup for DoD (Department of Defense) -Information Assurance compliance).

The provided text describes the acceptance criteria and supporting study for the AI features (Deep Resolve Boost, Deep Resolve Sharp, and Deep RxE) within the MAGNETOM Terra and MAGNETOM Terra.X devices.

Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

2. Automated unit-tests are set up to test the consistency of the generated output to a previously defined reference output.
3. During verification, the performance of the network is tested on a phantom against the ground truth with a maximal allowed NRMSE of 11% (for 2D network) and 8.7% (for 3D network).
4. The trained final network was used in the clinical study. (Implicit: Increases image homogeneity in a reproducible way on the receive profile, and images acquired with Deep RxE are rated better for image quality in the clinical study). | 1. During training, the loss, as the difference to a ground truth, is monitored and the training step with the lowest test loss is taken as the final trained network.
5. Automated unit-tests are set up to test the consistency of the generated output to a previously defined reference output.
6. During verification, the performance of the network is tested on a phantom against the ground truth with a maximal allowed NRMSE of 11% (11% for the 2D network and 8.7% for the 3D network were achieved).
7. The trained final network was used in the clinical study.
   The tests show that Deep RxE increases image homogeneity in a reproducible way on the receive profile. Images acquired with Deep RxE (DL bias field correction) are rated better for image quality than the ones acquired without it in the clinical study that was conducted. |

Note on Acceptance Criteria: The document directly states acceptance criteria for Deep RxE (e.g., NRMSE

Ask a specific question about this device

The MAGNETOM system is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These images and/or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

The MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

The subject device, MAGNETOM Cima.X Fit with software syngo MR XA61A, consists of new and modified software and hardware that is similar to what is currently offered on the predicate device, MAGNETOM Vida with syngo MR XA50A (K213693).

A high-level summary of the new and modified hardware and software is provided below:

For MAGNETOM Cima.X Fit with syngo MR XA61:

Hardware

New Hardware:
→ 3D Camera

Modified Hardware:

• → Host computers ((syngo MR Acquisition Workplace (MRAWP) and syngo MR Workplace (MRWP)).

• MaRS (Measurement and Reconstruction System).

• → Gradient Coil
• → Cover
• → Cooling/ACSC
• → SEP
• → GPA
• → RFCEL Temp
• → Body Coil
• → Tunnel light

Software

New Features and Applications:

• -> GRE_PC
• → Physio logging
• -> Deep Resolve Boost HASTE

• Deep Resolve Boost EPI Diffusion

• → Open Recon
• -> Ghost reduction (DPG)
• -> Fleet Ref Scan
• → Manual Mode
• → SAMER
• → MR Fingerprinting (MRF)1

Modified Features and Applications:

• → BEAT nav (re-naming only).

• myExam Angio Advanced Assist (Test Bolus).

• → Beat Sensor (all sequences).

• Stimulation monitoring

• -> Complex Averaging

I am sorry, but the provided text does not contain the acceptance criteria and the comprehensive study details you requested for the "MAGNETOM Cima.X Fit" device, particularly point-by-point information on a multi-reader multi-case (MRMC) comparative effectiveness study or specific quantitative acceptance criteria for its AI features like Deep Resolve Boost or Deep Resolve Sharp.

The document is a 510(k) summary for a Magnetic Resonance Diagnostic Device (MRDD), highlighting its substantial equivalence to a predicate device. While it mentions AI features and their training/validation, it does not provide the detailed performance metrics or study design to fully answer your request.

Here's what can be extracted based on the provided text, and where information is missing:

1. Table of Acceptance Criteria and Reported Device Performance:

The document mentions that the impact of the AI networks (Deep Resolve Boost and Deep Resolve Sharp) has been characterized by "several quality metrics such as peak signal-to-noise ratio (PSNR) and structural similarity index (SSIM)," and evaluated by "visual comparisons to evaluate e.g., aliasing artifacts, image sharpness and denoising levels" and "perceptual loss." For Deep Resolve Sharp, "an evaluation of image sharpness by intensity profile comparisons of reconstructions with and without Deep Resolve Sharp" was also conducted.

However, specific numerical acceptance criteria (e.g., PSNR > X, SSIM > Y), or the actual reported performance values against these criteria are not provided in the text. The document states that the conclusions from the non-clinical data suggest that the features bear an equivalent safety and performance profile to that of the predicate device, but no quantitative data to support this for the AI features is included in this summary.

• SSIM (implied to be high)
• Visual comparisons (e.g., absence of aliasing artifacts, good image sharpness, effective denoising levels) | Impact characterized by these metrics and visual comparisons. Claims of equivalent safety and performance profile to predicate device. No specific quantitative performance values (e.g., actual PSNR/SSIM scores) are reported in this document. |
  | Deep Resolve Sharp | - PSNR (implied to be high)
• SSIM (implied to be high)
• Perceptual loss
• Visual rating
• Image sharpness by intensity profile comparisons (reconstructions with and without Deep Resolve Sharp) | Impact characterized by these metrics, verified and validated by in-house tests. Claims of equivalent safety and performance profile to predicate device. No specific quantitative performance values are reported in this document. |

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective):

• Deep Resolve Boost:
  • Test Set Description: The text mentions that "the performance was evaluated by visual comparisons." It does not explicitly state a separate test set size beyond the validation data used during development. It implies the performance evaluation was based on the broad range of data covered during training and validation.
  • Data Provenance: Not specified (country of origin or retrospective/prospective). The data was "retrospectively created from the ground truth by data manipulation and augmentation."
• Deep Resolve Sharp:
  • Test Set Description: The text mentions "in-house tests. These tests include visual rating and an evaluation of image sharpness by intensity profile comparisons of reconstructions with and without Deep Resolve Sharp." Similar to Deep Resolve Boost, a separate test set size is not explicitly stated. It implies these tests were performed on data from the more than 10,000 high-resolution 2D images used for training and validation.
  • Data Provenance: Not specified (country of origin or retrospective/prospective). The data was "retrospectively created from the ground truth by data manipulation."

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

• Not specified. The document mentions "visual comparisons" and "visual rating" as part of the evaluation but does not detail how many experts were involved or their qualifications.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set:

• Not specified.

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 comparative effectiveness study is not mentioned in this document as being performed to establish substantial equivalence for the AI features. The document relies on technical metrics and visual comparisons of image quality to demonstrate equivalence.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

• The evaluation mentioned, using metrics like PSNR, SSIM, perceptual loss, and intensity profile comparisons, are indicative of standalone algorithm performance in terms of image quality. Visual comparisons and ratings would involve human observers, but the primary focus described is on the image output quality itself from the algorithm. However, no specific "standalone" study design with comparative performance metrics (e.g., standalone diagnostic accuracy) is detailed.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

• Deep Resolve Boost: "The acquired datasets (as described above) represent the ground truth for the training and validation." This implies the high-quality, full-data MRI scans before artificial undersampling or noise addition served as the ground truth. This is a technical ground truth based on the original acquired MRI data, not a clinical ground truth like pathology or expert consensus on a diagnosis.
• Deep Resolve Sharp: "The acquired datasets represent the ground truth for the training and validation." Similar to Deep Resolve Boost, this refers to technical ground truth from high-resolution 2D images before manipulation.

8. The sample size for the training set:

• Deep Resolve Boost:
  • TSE: more than 25,000 slices
  • HASTE: pre-trained on the TSE dataset and refined with more than 10,000 HASTE slices
  • EPI Diffusion: more than 1,000,000 slices
• Deep Resolve Sharp: on more than 10,000 high resolution 2D images.

9. How the ground truth for the training set was established:

• Deep Resolve Boost: "The acquired datasets (as described above) represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation and augmentation. This process includes further under-sampling of the data by discarding k-space lines, lowering of the SNR level by addition Restricted of noise and mirroring of k-space data."
• Deep Resolve Sharp: "The acquired datasets represent the ground truth for the training and validation. Input data was retrospectively created from the ground truth by data manipulation. k-space data has been cropped such that only the center part of the data was used as input. With this method corresponding low-resolution data as input and high-resolution data as output / ground truth were created for training and validation."

In summary, the document focuses on the technical aspects of the AI features and their development, demonstrating substantial equivalence through non-clinical performance tests and image quality assessments, rather than clinical efficacy studies with specific diagnostic accuracy endpoints or human-AI interaction evaluations.

Ask a specific question about this device

The MAGNETOM system is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These images and/or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

The MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

The subject device, MAGNETOM Cima.X with software syngo MR XA61A, consists of new and modified software and hardware that is similar to what is currently offered on the predicate device, MAGNETOM Vida with syngo MR XA50A (K213693).

A high-level summary of the new and modified hardware and software is provided below:

For MAGNETOM Cima.X with syngo MR XA61:

Hardware
New Hardware:
→ 3D Camera
Modified Hardware:

• → Host computers ((syngo MR Acquisition Workplace (MRAWP) and syngo MR Workplace (MRWP)).
• → MaRS (Measurement and Reconstruction System).
• → Gradient Coil
• → Cover
• → Cooling/ACSC
• → SEP
• → GPA
• → RFCEL Temp
• → Body Coil
• → Tunnel light

Software
New Features and Applications:

• -> GRE_PC
• → Physio logging
• -> Deep Resolve Boost HASTE
• → Deep Resolve Boost EPI Diffusion
• → Open Recon
• -> Ghost reduction (DPG)
• -> Fleet Ref Scan
• → Manual Mode
• → SAMER

Modified Features and Applications:

• → BEAT_nav (re-naming only).
• → myExam Angio Advanced Assist (Test Bolus).
• → Beat Sensor (all sequences).
• → Stimulation monitoring
• -> Complex Averaging

Additionally, the pulse sequence MR Fingerprinting (MRF) (K213805) is now available for the subject device MAGNETOM Cima.X with syngo MR XA61A.

The provided text is a 510(k) Summary for a medical device (MAGNETOM Cima.X) and outlines how the device, particularly its AI features, meets acceptance criteria through studies.

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implied by the performance characteristics used to evaluate the AI features. The reported device performance is presented in terms of quality metrics and visual evaluations.

• Peak Signal-to-Noise Ratio (PSNR)
• Structural Similarity Index (SSIM)
• Evaluated by visual comparisons to assess aliasing artifacts, image sharpness, and denoising levels. |
  | Deep Resolve Sharp | |
  | Image quality (e.g., sharpness) | Characterized by:
• Peak Signal-to-Noise Ratio (PSNR)
• Structural Similarity Index (SSIM)
• Perceptual loss
• Verified and validated by in-house tests, including visual rating and evaluation of image sharpness by intensity profile comparisons of reconstructions with and without Deep Resolve Sharp. |

2. Sample Sizes Used for the Test Set and Data Provenance

The document does not explicitly delineate a separate "test set" with a dedicated sample size after the training and validation phase for Deep Resolve Boost and Deep Resolve Sharp. Instead, it seems the "validation" mentioned in the context of training and validation data encompasses the evaluation of device performance.

• Deep Resolve Boost:

  • TSE: More than 25,000 slices (used for training and validation).
  • HASTE: Pre-trained on TSE dataset and refined with more than 10,000 HASTE slices (used for training and validation).
  • EPI Diffusion: More than 1,000,000 slices (used for training and validation).
  • Data Provenance: Retrospectively created from acquired datasets. The document does not specify the country of origin.

• Deep Resolve Sharp:

  • Sample Size: More than 10,000 high-resolution 2D images (used for training and validation).
  • Data Provenance: Retrospectively created from acquired datasets. The document does not specify the country of origin.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Their Qualifications

The document does not mention the use of experts to establish ground truth for the test set of the AI features. The "visual comparisons" and "visual rating" described are internal evaluations for feature performance but are not linked to expert-established ground truth for a formal test set described as such.

4. Adjudication Method for the Test Set

Not applicable, as no external expert-adjudicated test set is explicitly described for the AI features. The evaluations mentioned (visual comparisons, visual rating) appear to be internal.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No MRMC comparative effectiveness study is mentioned in the provided text for the AI features. The document focuses on the technical performance of the AI algorithms rather than their impact on human reader performance.

6. If a Standalone (i.e. algorithm only without human-in-the-loop performance) Was Done

Yes, the performance evaluation for Deep Resolve Boost and Deep Resolve Sharp appears to be standalone algorithm performance. The metrics (PSNR, SSIM, perceptual loss) and visual comparisons/ratings are related to the image quality produced by the algorithm itself, without direct assessment of human-in-the-loop performance.

7. The Type of Ground Truth Used

• Deep Resolve Boost: The acquired datasets (MRI raw data or images) were considered the "ground truth" for training and validation. Input data for the AI was then retrospectively created from this ground truth by data manipulation and augmentation (discarding k-space lines, lowering SNR, mirroring k-space data) to simulate different acquisition conditions.
• Deep Resolve Sharp: The acquired datasets (high-resolution 2D images) were considered the "ground truth" for training and validation. Low-resolution input data for the AI was retrospectively created from this ground truth by cropping k-space data, so the high-resolution data served as the output/ground truth.

8. The Sample Size for the Training Set

The document combines training and validation data, so the sample sizes listed in point 2 apply:

• Deep Resolve Boost:
  • TSE: More than 25,000 slices
  • HASTE: More than 10,000 HASTE slices (refined)
  • EPI Diffusion: More than 1,000,000 slices
• Deep Resolve Sharp: More than 10,000 high-resolution 2D images

9. How the Ground Truth for the Training Set Was Established

• Deep Resolve Boost: "The acquired datasets (as described above) represent the ground truth for the training and validation." This implies that the raw, original MRI data or images acquired under standard, full-sampling conditions were considered the reference. The AI was then trained to recover information from artificially degraded or undersampled versions of this ground truth.
• Deep Resolve Sharp: "The acquired datasets represent the ground truth for the training and validation." Similar to Deep Resolve Boost, the original high-resolution acquired 2D images were used as the ground truth. Low-resolution data was then derived from these high-resolution images to create the input for the AI, with the original high-resolution images serving as the target output (ground truth).

Ask a specific question about this device

The MAGNETOM system is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These images and/or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician vield information that may assist in diagnosis.

The MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

MAGNETOM Viato.Mobile with software syngo MR XA51A includes minor modified hardware compared to the predicate device. MAGNETOM Sola Fit with software syngo MR XA51A. A high level summary of the modified hardware is provided below:

Hardware
Modified Hardware

• Cover
  Other Modifications and / or Minor Changes
• Adaptations for installation in a mobile trailer
• MAGNETOM Viato.Mobile is a mobile MR system which enables the customers to relocate the MRI system to different locations and therefore provide imaging services where it is needed.

The provided text describes the 510(k) summary for the MAGNETOM Viato.Mobile device, focusing on its substantial equivalence to a predicate device. However, it does not contain information about acceptance criteria and a study specifically proving the device meets those criteria for software-driven performance aspects, nor does it include information about AI/ML models.

The document states: "No clinical study and no additional clinical tests were conducted to support substantial equivalence for the subject device." It primarily focuses on hardware modifications and compliance with general medical device standards.

Therefore, many of the requested details cannot be extracted from the provided text. Below is a summary of what can be inferred from the document and a clear indication of what information is missing.

Acceptance Criteria and Reported Device Performance

The document does not explicitly state quantitative "acceptance criteria" and "reported device performance" in the context of an AI/ML model for diagnostic accuracy. Instead, the "performance" discussed relates to the device's adherence to general safety and operational standards as a Magnetic Resonance Diagnostic Device (MRDD).

Table of Acceptance Criteria and Reported Device Performance (as inferred from the document regarding the device's overall functionality and safety):

Regarding the study that proves the device meets acceptance criteria:

The document explicitly states: "No clinical study and no additional clinical tests were conducted to support substantial equivalence for the subject device."

The assessment for substantial equivalence was based on:

1. Bench testing of modified hardware: Performed according to "Guidance for Submission of Premarket Notifications for Magnetic Resonance Diagnostic Devices."
2. Verification and validation (V&V) of modified hardware: Performed according to "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices."
3. Electrical safety and electromagnetic compatibility (EMC) testing of the complete system: Performed per IEC 60601-1-2.

The conclusion is that these non-clinical data demonstrate the device performs as intended and is substantially equivalent to the predicate device, the MAGNETOM Sola Fit (K221733).

Missing Information (Not found in the provided text):

1. Sample size used for the test set and the data provenance: Not applicable, as no performance study for diagnostic accuracy was conducted for an AI component. The tests were for hardware and system compliance.
2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable, as there was no test set requiring expert ground truth for diagnostic accuracy.
3. Adjudication method for the test set: Not applicable.
4. 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: Not applicable. The device is a Magnetic Resonance Diagnostic Device, not an AI-assisted diagnostic tool.
5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable.
6. The type of ground truth used (expert concensus, pathology, outcomes data, etc): Not applicable.
7. The sample size for the training set: Not applicable.
8. How the ground truth for the training set was established: Not applicable.

This device is primarily an MR hardware system with software for operation and image generation, not a device incorporating AI/ML for diagnostic interpretation. The substantial equivalence relies on proving the modified hardware and mobile integration retain the fundamental safety and performance characteristics of the predicate device, as demonstrated through engineering tests and adherence to recognized standards.

Ask a specific question about this device

The MAGNETOM MR system is indicated for use as a magnetic resonance diagnostic device (MRDD), which produces transverse, sagittal, coronal, and oblique cross sectional images that display the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images may also be produced. Depending on the region of interest, contrast agents may be used.

These images and the physical parameters derived from the images when interpreted by a trained physician yield information that may assist in diagnosis.

The subject device, MAGNETOM Free.Max with software syngo MR XA40A, is an 80 cm bore Magnetic Resonance Imaging system with an actively shielded 0.55T superconducting magnet. Which is the first 0.55T MRI system for clinical use in the U.S.

This FDA 510(k) summary for the MAGNETOM Free.Max MRI system focuses on demonstrating substantial equivalence to a predicate device rather than providing specific acceptance criteria for a new AI/CADe algorithm. Therefore, much of the requested information regarding AI performance metrics, sample sizes for test/training sets, expert qualifications, and ground truth establishment is not present in this document.

However, I can extract the information that is available:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly define acceptance criteria in terms of numerical thresholds for specific performance metrics (e.g., sensitivity, specificity for a diagnostic task). Instead, the performance testing aims to demonstrate equivalence in image quality and safety to the predicate device.

2. Sample size used for the test set and the data provenance

• Test Set (for PNS study): 12 individuals
• Data Provenance: Not explicitly stated, but the PNS study was a "clinical study" suggesting prospective data collection. The software verification and validation would likely use a mix of internally generated and potentially simulated data. Sample clinical images would be from human subjects but their precise origin isn't detailed beyond being "sample clinical images."

3. 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 describe the establishment of a "ground truth" by experts in the context of an algorithmic diagnostic performance study. The images are "interpreted by a trained physician" as per the Indications for Use, which is general clinical practice, not a specific ground truth establishment for algorithm evaluation.

4. Adjudication method for the test set

• Not applicable. No expert adjudication method is described for an algorithmic performance evaluation.

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 comparative effectiveness study involving human readers with and without AI assistance is mentioned. This submission is for the MRI system itself, not an AI-powered diagnostic aid that assists human readers.

6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done

• Not applicable. The "software verification and validation" would assess the software's functional performance, but not in the context of a standalone diagnostic algorithm providing a clinical output that would typically be evaluated for sensitivity/specificity. The Deep Resolve Gain and Deep Resolve Sharp features hint at image processing algorithms, but their standalone diagnostic performance is not presented.

7. The type of ground truth used

• For the PNS study, the "ground truth" would be the physiological response of the individuals.
• For image quality assessments, the ground truth is subjective visual assessment and objective metrics (SNR, uniformity) compared against engineering specifications and predicate device performance.
• For software verification and validation, the ground truth is adherence to design specifications and expected functional behavior.
• No "expert consensus, pathology, or outcomes data" ground truth is described in the context of validating a diagnostic algorithm's performance against clinical findings.

8. The sample size for the training set

• Not specified. The document refers to "Deep Resolve Gain" and "Deep Resolve Sharp" which are likely AI-based image processing features. However, the details of their training data (sample size, origin, ground truth) are not provided. The listed clinical publications for these features (e.g., "Residual Dense Network for Image Super-Resolution") suggest they are based on deep learning techniques that would require training data.

9. How the ground truth for the training set was established

• Not specified. As noted above, details about training data for any potential AI components (like Deep Resolve) and their ground truth are not included in this summary.

Summary of what the document indicates about the device:

This 510(k) submission for the MAGNETOM Free.Max MRI system focuses on demonstrating substantial equivalence to an existing predicate device (MAGNETOM Sempra) by:

• Comparing technological characteristics (hardware and software).
• Performing non-clinical tests (sample clinical images, image quality assessments, bench tests for SNR/uniformity/heating, and general software V&V) to ensure the new device performs effectively and safely in a manner equivalent to the predicate.
• Conducting a small clinical study on Peripheral Nerve Stimulation (PNS) effects for safety.
• Referencing clinical publications for various new software features, implying that the underlying scientific principles and expected clinical utility of these features are generally accepted.

The document does not detail the validation of a specific AI/CADe diagnostic algorithm with acceptance criteria related to clinical diagnostic performance metrics. If "Deep Resolve Gain" and "Deep Resolve Sharp" involve AI, their validation is presented as part of overall system performance and image quality rather than as a standalone diagnostic aid.

Ask a specific question about this device

Your MAGNETOM system is indicated for use as a magnetic resonance diagnostic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional images, spectroscopic images and/or spectra, and that displays the internal structure and/or function of the head, body, or extremities. Other physical parameters derived from the images and/or spectra may also be produced. Depending on the region of interest, contrast agents may be used. These images and or spectra and the physical parameters derived from the images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

Your MAGNETOM system may also be used for imaging during interventional procedures when performed with MR compatible devices such as in-room displays and MR Safe biopsy needles.

MAGNETOM Vida, MAGNETOM Sola, MAGNETOM Lumina, MAGNETOM Altea with software syngo MR XA31A includes new and modified hardware and software compared to the predicate device, MAGNETOM Vida with software syngo MR XA20A.

This document describes the Siemens MAGNETOM MR system (various models) with syngo MR XA31A software, and it does not describe an AI device. The information provided is a 510(k) summary for a Magnetic Resonance Diagnostic Device (MRDD). The "Deep Resolve Sharp" and "Deep Resolve Gain" features are mentioned as using "trained convolutional neuronal networks" but the document does not provide details on acceptance criteria or studies specific to the AI components as requested.

Therefore, many of the requested items (e.g., sample sizes for training/test sets for AI, expert consensus for ground truth, MRMC studies) cannot be extracted from this document because it is primarily focused on the substantial equivalence of the overall MR system and its general technological characteristics, not a specific AI algorithm requiring detailed performance studies against a clinical ground truth.

However, I can extract the available information, especially concerning the "Deep Resolve Sharp" and "Deep Resolve Gain" features, and note where the requested information is not present.

Here's the breakdown of available information, with specific answers to your questions where possible:

1. A table of acceptance criteria and the reported device performance

The document does not specify quantitative acceptance criteria for the "Deep Resolve Sharp" or "Deep Resolve Gain" features, nor does it present a table of reported device performance metrics for these features in the context of clinical accuracy or diagnostic improvement specifically. The performance testing mentioned is general for the entire system ("Image quality assessments," "Performance bench test," "Software verification and validation"), concluding that devices "perform as intended and are thus substantially equivalent."

2. Sample sizes used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Test Set Sample Size: Not explicitly stated for specific features like "Deep Resolve Sharp" or "Deep Resolve Gain." The document broadly mentions "Sample clinical images" were used for "Image quality assessments."
• Data Provenance (Country/Retrospective/Prospective): Not specified in the document.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)

Not specified. The document states "Image quality assessments by sample clinical images" and that the "images...when interpreted by a trained physician yield information that may assist in diagnosis," but it does not detail the number or qualifications of experts involved in these assessments for specific software features or for establishing ground truth for any AI component.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not specified.

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

An MRMC study was not described for the "Deep Resolve Sharp" or "Deep Resolve Gain" features or any other AI component. The document references clinical publications for some features (e.g., Prostate Dot Engine, GRE_WAVE, SVS_EDIT) but these are general publications related to the underlying clinical concepts or techniques, not comparative effectiveness studies of the system's AI features versus human performance. The statement "No additional clinical tests were conducted to support substantial equivalence for the subject devices" reinforces this.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

While "Deep Resolve Sharp" and "Deep Resolve Gain" involve "trained convolutional neuronal networks," the document does not describe standalone performance studies for these algorithms. Their inclusion is framed as an enhancement to the overall MR system's image processing capabilities, rather than a separate diagnostic AI tool. The stated purpose of Deep Resolve Sharp is to "increases the perceived sharpness of the interpolated images" and Deep Resolve Gain "improves the SNR of the scanned images," both being image reconstruction/enhancement features.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

Not specified for any AI-related features. For general image quality assessment, the "trained physician" is mentioned as interpreting images to assist in diagnosis, implying clinical interpretation, but no formal ground truth establishment process is detailed.

8. The sample size for the training set

Not specified for the "trained convolutional neuronal networks" used in "Deep Resolve Sharp" or "Deep Resolve Gain."

9. How the ground truth for the training set was established

Not specified.

Ask a specific question about this device