Search Results

The SIGNA Architect system is a whole body magnetic resonance scanner designed to support high signal-to-noise ratio, and short scan times. It is indicated for use as a diagnostic imaging device to produce axial, sagittal, coronal, and oblique images, spectroscopic images, parametric maps, and/or spectra, dynamic images of the structures and/or functions of the entire body, including, but not limited to, head, neck, TMJ, spine, breat, abdomen, pelvis, joints, prostate, blood vessels, and musculoskelatal regions of the region of interest being imaged, contrast agents may be used.

The images produced by the SIGNA Architect system reflect the spatial distribution or molecular environment of nuclei exhibiting magnetic resonance. These images and/or spectra when interpreted by a trained physician vield information that may assist in diagnosis.

SIGNA Architect is a whole body magnetic resonance scanner designed to support high resolution, high signal-to-noise ratio, and short scan times. The system features a superconducting magnet. The data acquisition system accommodates up to 128 independent receive channels in various increments and multiple independent coil elements per channel during a single acquisition series. Each system uses a combination of time varying magnetic fields (gradients) and RF transmissions to obtain information regarding the density and position of elements exhibiting magnetic resonance. Each system can imaqe in the saqittal, coronal, axial, oblique, and double oblique planes, using various pulse sequences and reconstruction algorithms.

This submission is prompted by the introduction of two new software features called HyperSense 2.0 and Star onto SIGNA Architect. HyperSense 2.0 is an acceleration technique based on sparse data compressibility allowing scan time reduction while maintaining SNR efficiency. Star is a motion-robust, free-breathing imaging technique. HyperSense 2.0 is a modification to the previously cleared HyperSense, while Star is a technique that can be used with the previously cleared DISCO feature. Both HyperSense and DISCO are listed above as reference devices along with their associated 510(k) submission numbers.

The addition of both the HyperSense 2.0 and Star features involved modifications to the SIGNA Architect system software. There were no changes from either of these features that were related to the system's hardware components.

The provided text describes a 510(k) premarket notification for a medical device, the SIGNA Architect, a Magnetic Resonance (MR) system with new software features (HyperSense 2.0 and Star). Here's a breakdown of the acceptance criteria and study proving the device meets them, based on the information provided:

Disclaimer: The provided document is a 510(k) summary, which is a high-level overview. It does not contain detailed information about the specific acceptance criteria, statistical methodologies, or all aspects of the studies that would be present in the full submission. Therefore, some sections below will indicate "Not explicitly stated in the provided document."

Acceptance Criteria and Device Performance

The core acceptance criterion for a 510(k) submission is Substantial Equivalence (SE) to a legally marketed predicate device. This means the new device is as safe and effective as the predicate, and does not raise new questions of safety and effectiveness.

The document indicates that studies were performed to demonstrate that the new features (HyperSense 2.0 and Star) do not negatively impact image quality or diagnostic utility compared to the predicate/existing techniques.

Table of Acceptance Criteria and Reported Device Performance:

Study Details

The document mentions two main types of studies: non-clinical and clinical. The clinical evaluation focuses on the new features: HyperSense 2.0 and Star.

1. Sample Size Used for the Test Set and Data Provenance:

• HyperSense 2.0: "The images involved were generated using 3 different reconstruction techniques across different anatomies."
  • Sample Size: Not explicitly stated (e.g., number of patients/cases, number of images).
  • Data Provenance: Not explicitly stated (e.g., country of origin, retrospective or prospective).
• Star: "Images from the assessment are provided."
  • Sample Size: Not explicitly stated (e.g., number of patients/cases, number of images).
  • Data Provenance: Not explicitly stated (e.g., country of origin, retrospective or prospective).

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:

• HyperSense 2.0: "Radiologists were asked to evaluate side-by-side image quality of the HyperSense 2.0 images compared to the predicate."
  • Number of Experts: "Radiologists" (plural), but specific number not stated.
  • Qualifications: Not explicitly stated (e.g., years of experience, subspecialty).
• Star: "Images produced by Star were judged to be of sufficient quality for diagnostic use by a a U.S. Board Certified radiologist."
  • Number of Experts: "a U.S. Board Certified radiologist" (singular).
  • Qualifications: "U.S. Board Certified radiologist." (Years of experience or subspecialty not stated).

3. Adjudication Method (for the test set):

• HyperSense 2.0: "Radiologists were asked to evaluate side-by-side image quality of the HyperSense 2.0 images compared to the predicate." This suggests individual evaluation rather than a formal adjudication process between multiple readers.
  • Method: Not explicitly stated beyond individual reader evaluation of side-by-side images. No mention of 2+1, 3+1, or consensus.
• Star: "Images produced by Star were judged to be of sufficient quality for diagnostic use by a U.S. Board Certified radiologist."
  • Method: Single reader evaluation. No adjudication described.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

• HyperSense 2.0: The description "An external reader evaluation study was performed" and "Radiologists were asked to evaluate side-by-side image quality" suggests a multi-reader study, but it's not explicitly labeled as a formal MRMC study. The details provided are insufficient to confirm the rigor of a full MRMC design (e.g., statistical analysis of reader performance differences).
  • Effect Size of Human Readers Improve with AI vs. without AI assistance: This specific metric is not applicable here as the described studies focus on image quality assessment of a new image acquisition/reconstruction technique, not directly on AI assisting human readers in a diagnostic task for a specific condition. The "AI" implied (HyperSense 2.0 and Star) are image processing algorithms, not diagnostic AI systems assisting in interpretations.

5. If a Standalone (i.e. algorithm only, without human-in-the-loop performance) was done:

• The non-clinical testing for both features would implicitly include standalone performance evaluation of the algorithms (e.g., technical measures of SNR, resolution, artifact reduction), but the document does not elaborate on these specific "standalone" metrics or a formal standalone study results. The clinical evaluations do involve human assessment of the images produced by the algorithms.

6. The Type of Ground Truth Used:

• The "ground truth" for these studies appears to be expert consensus/opinion on image quality and diagnostic sufficiency.
  • For HyperSense 2.0, the radiologists' evaluation of "overall image quality and uniformity" served as the basis for acceptance.
  • For Star, the "U.S. Board Certified radiologist's" judgment of "sufficient quality for diagnostic use" served as the basis for acceptance.
  • There is no mention of pathology, long-term outcomes data, or other objective ground truths beyond expert interpretation of the images themselves.

7. The Sample Size for the Training Set:

• This information is Not explicitly stated in the provided document. The document details the testing of the software features, but not the development or training set size (if algorithms involved machine learning).

8. How the Ground Truth for the Training Set was Established:

• This information is Not explicitly stated in the provided document. As the training set size isn't mentioned, neither is its ground truth establishment.

Ask a specific question about this device

The SIGNA Architect, SIGNA Artist, Discovery MR750 3.0T, Discovery MR450 1.5T, Discovery MR750w 3.0T and the Optima MR450w 1.5T systems are whole body magnetic resonance scanners designed to support high signal-to-noise ratio, and short scan times. It is indicated for use as a diagnostic imaging device to produce axial, sagittal. coronal, and oblique images, spectroscopic images, parametric maps, and/or spectra, dynamic images of the structures and/or functions of the entire body, including, but not limited to, head, neck, TMI, spine, breast, heart, abdomen, pelvis, joints, prostate, blood vessels, and musculoskeletal regions of the body. Depending on the region of interest being imaged, contrast agents may be used.

The images produced by the SIGNA Architect, SIGNA Artist, Discovery MR750 3.0T, Discovery MR450 1.5T, Discovery MR750w 3.0T and the Optima MR450w 1.5T systems reflect the spatial distribution or molecular environment of nuclei exhibiting magnetic resonance. These images and/or spectra when interpreted by a trained physician vield information that may assist in diagnosis.

The Discovery MR750 3.0T, Discovery MR450 1.5T, Discovery MR750w 3.0T, Optima MR450w 1.5T, SIGNA Architect and SIGNA Artist systems are whole body magnetic resonance scanners designed to support high resolution, high signal-to-noise ratio, and short scan times. The systems each feature a superconducting magnet. The data acquisition system accommodates up to 128 independent receive channels in various increments and multiple independent coil elements per channel during a single acquisition series. Each system uses a combination of time varying magnetic fields (gradients) and RF transmissions to obtain information regarding the density and position of elements exhibiting magnetic resonance. Each system can image in the sagittal, coronal, axial, oblique, and double oblique planes, using various pulse sequences and reconstruction algorithms. The Discovery MR750 3.0T, Discovery MR450 1.5T, Discovery MR750w 3.0T, Optima MR450w 1.5T, SIGNA Architect. SIGNA Artist systems are designed to conform to NEMA DICOM standards (Digital Imaging and Communications in Medicine).

The original description hasn't changed from predicate devices (K160618), other than reflecting the additional receive channels available.

The modifications to these systems include the MAGIC DWI and CardioMaps software features, delivered via the DV26 program. The proposed software features will be ported over to other GE Healthcare MR systems based on appropriate design controls and evaluation of the change in accordance with FDA's Guidance—Deciding When to Submit a 510(k) for a Change to an Existing Device.

This document describes the premarket notification (510(k)) for GE Medical Systems' SIGNA Architect, SIGNA Artist, Discovery MR750 3.0T, Discovery MR450 1.5T, Discovery MR750w 3.0T and the Optima MR450w 1.5T Magnetic Resonance (MR) diagnostic devices. The submission focuses on the addition of MAGIC DWI (Diffusion-Weighted Imaging) and CardioMaps software features.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state specific quantitative acceptance criteria or performance metrics for the MAGIC DWI and CardioMaps software features in a table format. Instead, it indicates that testing was completed with "passing results per the pass/fail criteria defined in the test cases."

Implicit Acceptance Criteria (inferred from the document):

• Safety and Effectiveness: The primary acceptance criterion is that the modified software features (MAGIC DWI and CardioMaps) are "as safe and effective as the predicate" devices and do "not raise different questions of safety and effectiveness."
• Compliance with Standards: The software features must comply with voluntary standards: AAMI/ANSI 62304, AAMI/ANSI ES60601-1, and IEC 60601-2-33.
• Acceptable Performance: Phantom testing for both software features must demonstrate "acceptable performance."

Reported Device Performance:

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

• Test Set Sample Size: The document does not specify a numerical sample size for either the phantom testing or the clinical images. It generically refers to "phantom testing" and "sample clinical images."
• Data Provenance: Not explicitly stated. For phantom testing, it's typically controlled laboratory conditions. For clinical images, it's not mentioned whether they are retrospective or prospective, nor their country of origin.

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

The document does not provide this information.
The summary states that images and/or spectra are interpreted by a "trained physician," but it doesn't detail the number or qualifications of experts involved in establishing ground truth for the specific performance evaluation of the new software features.

4. Adjudication Method for the Test Set

The document does not specify an adjudication method.
It states that "passing results per the pass/fail criteria defined in the test cases" were achieved for phantom testing. For clinical images, it mentions they are "interpreted by a trained physician," implying clinical judgment, but no formal adjudication process (like 2+1 or 3+1) is described for the evaluation presented in this summary.

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

The document does not indicate that an MRMC comparative effectiveness study was performed.
The evaluation relies on compliance with standards, phantom testing, and presentation of sample clinical images to demonstrate "substantial equivalence" rather than a comparative effectiveness study measuring human reader improvement with AI assistance. The software features are enhancements to image acquisition and processing, not explicitly AI-assisted diagnostic tools in the context of comparative reading studies.

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

While the software features (MAGIC DWI and CardioMaps) represent algorithm-only additions, the document emphasizes that the "images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis."
The "phantom testing" and quality assurance measures (e.g., unit-level, integration, performance, safety testing) can be considered standalone evaluations of the algorithms' output quality and adherence to specifications. However, the ultimate "performance" in the diagnostic context is tied to physician interpretation. The regulatory focus here is on the system producing diagnostically useful images, not on an algorithm making a standalone diagnosis.

7. The Type of Ground Truth Used

• For Phantom Testing: The ground truth would typically be established by known physical properties or measurements of the phantom itself. The "pass/fail criteria" would be based on expected quantitative accuracy, image quality, or signal properties against these known values.
• For Clinical Images: The document mentions "images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis." This implies that the effectiveness in a clinical setting is ultimately judged by expert clinical interpretation, but it does not specify a formal "ground truth" (e.g., pathology, surgical findings, long-term outcomes) used to validate the clinical utility of the specific new software features. It's more about demonstrating that the images produced can be interpreted by a physician to assist diagnosis.

8. The Sample Size for the Training Set

The document does not provide any information regarding a training set sample size. This is likely because the referenced software features are defined as modifications to existing MR systems, and while they involve algorithms, the summary doesn't describe them as machine learning models that require distinct "training sets" in the typical sense. The development process described (risk analysis, requirements reviews, design reviews, various levels of testing) is a standard software engineering approach.

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

As no training set is mentioned for machine learning, information on how its ground truth was established is not applicable or provided in this document.

Ask a specific question about this device