FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

Your MAGNETOM system is indicated for use as a magnetic device (MRDD) that produces transverse, sagittal, coronal and oblique cross sectional 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 inages 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.

Device Description

MAGNETOM Vida with software syngo MR XA11A includes new and modified hardware and software compared to the predicate device, MAGNETOM Vida with syngo MR XA10A. A high level summary of the new and modified features is provided below:

Hardware
New Hardware

New coils:
BM Body 12 |
BM Spine 24 |
| Head/Neck 16
-Head 32 MR Coil 3T
Other components:
camera —
computer
Multi-Channel Interface —

Modified Hardware

Main components such as 32 independent RF channels -
-Other components such as Tx-Box / RF filter plate / transmit system

Software
New Features and Applications

GOKnee3D (examination comprising the AutoAlign knee localizer and two SPACE with CAIPIRINHA sequences to support fast high-resolution 3D exams of the knee)
SPACE with CAIPIRINHA (3D SPACE pulse sequence type with the iPAT mode CAIPIRINHA)
GOBrain (brain examination in short acquisition time)
GOBrain+ (adaptation of GOBrain pulse sequences)
| MR Breast Biopsy (supports planning and execution of MR guided breast biopsies and wire localizations)
| MRSim / Synthetic CT (provides MR pulse sequences for the creation of Synthetic CT images based on the MR image input)
Cardiac Dot Flow Add-In (extension of Cardiac Dot Engine to support blood flow measurements)
PCASL mode (extension of ASL pulse sequence types by a new blood labeling mode)
SMS in TSE (Simultaneous Multi Slice (SMS) support for TSE)

Modified Features and Applications

| SliceAdjust (the framework support was extended to include additional pulse sequence types)
RetroGating (Compressed Sensing Cardiac Cine acquisitions which split the data acquisition over multiple heartbeats can now be configured to perform complete sampling of the cardiac cycle without prior definition of an acquisition window. Combination with arrhythmia rejection is possible.)
iPAT / TSE Reference Scan (Changes in the TSE, FAST TSE and TSE DIXON pulse sequence types includes the possibility to use a reference scan "TSE/Separate" for GRAPPA acquisition and reconstruction)
Care Bolus in Angio Dot Engine (workflow support for bolus administration (bolus detection))
MRCP in SPACE (improvement of the image quality for MR Cholangiopancreatography (MRCP) acquisitions based on the SPACE pulse sequence type)
MR Elastography:
Replacement of existing masking by masking performed on the prescan images used within the prescan/normalize (PSN) functionality.
Optimization of pulse sequence type timing. |
Changes in MEG time period (no longer fixed to the wavelength of the | MEG and also implementation of a reduced MEG period)
Respiratory Sensor Support (additional support for respiratory triggered measurements is provided in several SE-, GRE- and EPI-based pulse sequence types)

Modified (general) Software / Platform

ー Single and dual monitor workflow (In the single monitor setup the features of the LHS monitor and RHS monitor are provided on separate tab-cards)
Touch positioning (Select&GO 2.0) (extension to additional body area positions when dedicated coils are plugged in)
Dot Cockpit (additional features for handling of scan pulse sequences and offline Dot Cockpit)
MR View&GO (Addition of Mosaic View (view mode to scroll through dimensions | instead of space) and 4D Movie Toolbar (movie toolbar to navigate the 4th dimension))

Other Modifications and / or Minor Changes

teamplay Protocols Interface (interface to support external pulse sequences | management systems)
Unilateral Hip (added in Large Joint Dot Engine) (user workflow optimized, since information/settings are taken from the patient registration)
GRE RefScan (external GRE RefScan has been extended to multiple pulse sequence types)
Asymmetric saturation pulses (support for regional saturation with an asymmetric shape has been added for BOLD imaging)
CP Mode modification ("RF Transmit Mode" is provided as part of the patient registration based on IEC 60601-2-33)
SPAIR FatSat (new "SPAIR Breast" mode in several pulse sequence types and extension of "Abdomen&Pelvis" and "Thorax" modes)
Compressed Sensing GRASP-VIBE (improvement of SPAIR fat saturation performance)
MAGNETOM RT Pro Edition marketing bundle (extension of the bundle)
Siemens "BioMatrix" (extension with additional components)

AI/ML Overview

The provided text is a 510(k) Summary for the Siemens MAGNETOM Vida MRI system (K181433). It describes the device, its intended use, and compares it to a predicate device (MAGNETOM Vida with syngo MR XA10A). However, this document primarily focuses on establishing substantial equivalence based on non-clinical testing and adherence to standards, rather than clinical performance studies with acceptance criteria in the typical sense for AI/CADe devices.

Therefore, many of the requested details regarding acceptance criteria, clinical study design (sample size, expert qualifications, adjudication, MRMC studies, standalone performance), and ground truth establishment (especially for AI/ML models) are not present in this document. This is because this submission is for an MRI system, not an AI/CADe device. It focuses on hardware and software modifications of a diagnostic imaging machine, not on an algorithm that interprets images.

Based on the provided text, here's what can be inferred:

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

The document discusses "performance testing" but does not provide specific quantitative acceptance criteria or detailed reported performance in a table format as might be expected for an AI/CADe device. Instead, the "acceptance" is qualitative:

Acceptance Criteria (Inferred from Text)	Reported Device Performance (Inferred from Text)
New coils perform as intended.	Sample clinical images were taken and deemed satisfactory.
New/modified software features and algorithms perform as intended.	Image quality assessments were completed. In some cases, comparison to predicate device features showed equivalent image quality.
Software development adheres to medical device software standards (IEC 62304:2006).	Software verification and validation testing was completed in accordance with FDA guidance.
System performance aligns with FDA guidance for Magnetic Resonance Diagnostic Devices.	Performance tests were completed in accordance with FDA guidance document.
Device safety and effectiveness are established through risk management (ISO 14971:2007) and adherence to other recognized standards (e.g., IEC 60601 series, NEMA).	Risks are controlled through hardware/software development, testing, and labeling. Compliance with listed standards is affirmed.
Device is substantially equivalent to the predicate.	"The results from each set of tests demonstrate that the device performs as intended and is thus substantially equivalent to the predicate device to which it has been compared."

2. Sample sized used for the test set and the data provenance:

Test Set Sample Size: Not explicitly stated. The document mentions "sample clinical images" were taken for the new coils and software features, but no specific number of patients or images is given.
Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
Retrospective or Prospective: Not specified.

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

Number of Experts: Not specified.
Qualifications of Experts: The device's output is "interpreted by a trained physician," implying that physicians are involved in assessing the images, but their specific role in establishing "ground truth" for the non-clinical tests is not detailed. For this type of MRI system submission, ground truth isn't established in the same way as for an AI interpretation algorithm. The "truth" is the physical output of the MRI system.

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

Not applicable/Not specified. This level of detail on ground truth adjudication is typically for AI/CADe clinical studies, not MRI system performance tests focused on image quality and safety.

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. The document explicitly states: "No clinical tests were conducted to support substantial equivalence for the subject device". This is not an AI-assisted reading device, but a diagnostic imaging system.

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

Not applicable. This is an MRI device, not an AI algorithm. "Performance tests" were done on the device itself and its components (e.g., image quality assessments).

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

For an MRI system, the "ground truth" for non-clinical testing refers to the physical and technical performance of the device (e.g., image clarity, signal-to-noise ratio, spatial resolution, adherence to safety limits). It is not about diagnostic accuracy against a clinical ground truth like pathology. The comparison is made against prior versions/predicate devices and established industry standards for image quality and safety.

8. The sample size for the training set:

Not applicable. This document does not describe an AI/ML algorithm that requires a training set.

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

Not applicable. As above, no AI/ML training set is discussed.

In summary: The provided document is a 510(k) summary for a Magnetic Resonance Diagnostic Device (MRDD), the MAGNETOM Vida MRI system. Its purpose is to demonstrate substantial equivalence to a legally marketed predicate device based on non-clinical performance testing (e.g., image quality assessments, software verification/validation) and conformity to recognized standards (e.g., IEC, ISO, NEMA). It explicitely states that no clinical tests were conducted for this submission. Therefore, the detailed requirements for AI/CADe device performance studies (like MRMC, training/test set ground truth, expert adjudication, etc.) are not addressed in this document because they are not relevant to this specific type of medical device submission.

Summary

Regulation Number and Section

§ 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.