Search Results

HyperBand is a software option intended for use on GE MR 1.5T and 3T Systems. HyperBand consists of an acquisition and reconstruction technique allowing simultaneous excitation of multiple slices at multiple locations to accelerate imaging acquisition times or increase slice coverage without increasing scan time. HyperBand is indicated for echo-planar imaging (EPI) of the head and breast. HyperBand is indicated for imaging all patients not otherwise contraindicated for MRI examination.

HyperBand is a clinical software application which is offered as an optional feature for GE Healthcare's 1.5T and 3T MR systems. It provides a reduction in scan time by simultaneously exciting multiple slices at multiple locations. It can lead to higher acceleration reduction factors when combined to other methods of parallel imaging. The benefits of HyperBand acceleration include enhancements on productivity and patient experience by shortened breath holds, increased anatomy coverage and higher resolution image acquisition.

The provided document is a 510(k) Premarket Notification Submission for a device called "HyperBand" by GE Medical Systems. This device is a software option for GE MR 1.5T and 3T Systems, intended to accelerate imaging acquisition times or increase slice coverage.

The document discusses the substantial equivalence of HyperBand to a predicate device (ASSET, K012970) and presents performance data to support this claim. However, it does not provide a table of acceptance criteria with reported device performance in terms of specific metrics like sensitivity, specificity, or accuracy. It also does not detail a study that defines explicit acceptance criteria with numerical targets.

Based on the information available, here's a breakdown of what can be extracted and what is missing:

1. Table of Acceptance Criteria and Reported Device Performance

Critique: The provided document uses qualitative statements like "maintain the same imaging performance" and "as safe and effective as the predicate" rather than specific, quantitative acceptance criteria with numerical targets (e.g., "SNR must be within X% of predicate," "image quality rated as diagnostic by Y% of radiologists"). This is typical for a 510(k) submission where "substantial equivalence" is the goal, often relying on comparison to an already cleared device.

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

• Test Set Sample Size: Not explicitly stated. The document mentions "Internal scans were conducted as part of validation for workflow and image quality for the addition of the new features." It does not specify the number of scans or patients included in this validation.
• Data Provenance: "Internal scans were conducted." This implies the data was generated within GE Healthcare, possibly at their facilities or in collaboration with internal sites. It is prospective in the sense that these scans were conducted for the purpose of validating the new features. The country of origin is not specified, but given GE Healthcare's operations, it could be the US or international.

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: Not specified. The document mentions "clinical results" and "image quality," implying assessment by medical professionals, but their specific roles, experience, or specialty (e.g., radiologist) are not detailed.

4. Adjudication method for the test set

• Adjudication Method: Not specified. Since the number of experts is not stated, it's impossible to determine if a method like 2+1, 3+1, or any other consensus-based adjudication was used.

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

• MRMC Study: No, an MRMC comparative effectiveness study involving human readers with and without AI assistance was not explicitly mentioned or performed. HyperBand is described as an acquisition and reconstruction technique to accelerate imaging, not an AI-assisted diagnostic tool that helps human readers improve their diagnostic performance. The focus is on maintaining image quality while improving scan efficiency.

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

• Standalone Performance: Not directly applicable in the terms of a diagnostic algorithm. HyperBand is a software option that modifies the MR acquisition and reconstruction process. Its "performance" is inherent in the generated images. The non-clinical tests (SNR testing) and clinical validation were effectively "standalone" tests of the system's output (image quality) but not in the context of diagnostic accuracy without human interpretation. The system (MR scanner with HyperBand) produces the images, and the validation checks if those images are diagnostically equivalent.

7. The type of ground truth used

• Type of Ground Truth: The document implies that the ground truth for image quality and workflow was established through comparative assessment against images acquired using the predicate device (ASSET). The "clinical results demonstrated that HyperBand maintain the same imaging performance results as its predicate devices." This suggests that image quality was judged to be comparable or equivalent, likely by expert review, to images obtained with an already accepted method. It's not pathology, or independent outcomes data, but rather a performance comparison against an established baseline.

8. The sample size for the training set

• Training Set Sample Size: Not applicable/not provided. HyperBand is described as an acquisition and reconstruction technique, not a machine learning algorithm that requires a "training set" in the conventional sense (i.e., for learning diagnostic patterns from annotated data). While software development involves testing and iterative refinement, the concept of a distinct 'training set' for an AI model is not mentioned for this device.

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

• Ground Truth for Training Set: Not applicable, as no training set (for AI/ML models) is mentioned.

Summary of Study:

The study referenced in the document is an internal validation conducted by GE Healthcare. Its primary goal was to demonstrate substantial equivalence of HyperBand to its predicate device (ASSET) in terms of image quality and overall safety and effectiveness. This was achieved through:

• Non-clinical tests: Including Signal to Noise Ratio (SNR) testing (with and without HyperBand) and compliance with voluntary standards (AAMI/ANSI 62304, AAMI/ANSI ES60601-1, IEC 60601-2-33, NEMA PS3.1-3.18 for DICOM conformance).
• Clinical tests (internal scans): These were conducted "as part of validation for workflow and image quality" to show that HyperBand maintains "the same imaging performance results as its predicate devices."

The study did not involve formal, external clinical trials with large patient cohorts or multi-reader studies to establish diagnostic performance against a definitive ground truth like pathology for specific conditions. Instead, it focused on demonstrating that the technical performance and resulting image quality of the accelerated acquisition technique were comparable to existing, cleared technology, thereby not raising new questions of safety or effectiveness.

Ask a specific question about this device

The GE Signa Infinity line of products are whole body magnetic resonance scanners designed to support high resolution, high signal-to-noise ratio and short scan times. The Signa Infinity System is indicated for use as a diagnostic imaging device to produce transverse, sagittal, coronal and oblique images, spectroscopic images, and/or spectra, dynamic images of the internal structures and organs of the entire body, including, but not limited to, head, neck, TMJ, spine, breast, heart, abdomen, pelvis, joints, prostate, blood vessels, and musculoskeletal regions of the images produced by the Signa Infinity systems reflect the spatial distribution of protons (hydrogen nuclei) exhibiting magnetic resonance. These images and/or spectra when interpreted by a trained physician yield information that may assist in diagnosis.

The Signa Infinity Magnetic Resonance System is a modification to the Signa TwinSpeed Magnetic Resonance System (K012200) which utilizes a superconducting magnet to acquire 2D single-slice and multi-slice, and 3D volume images. The Signa Infinity Magnetic Resonance System features a superconducting magnet operating at 1.5T. The system can image in the sagittal, coronal, axial, oblique and double oblique planes, using various pulse sequences. Images are acquired and reconstructed using 2D and 3D Fourier transformation techniques. The system is intended for high resolution anatomical applications and short scan times. Previously cleared software options, coils, and other accessories may · be used with the Signa Infinity Magnetic Resonance System.

Here's an analysis of the provided text regarding the acceptance criteria and study for the GE Signa Infinity MRI System:

Based on the provided 510(k) summary, the GE Signa Infinity Magnetic Resonance System is presented as a modification of an existing device, the Signa TwinSpeed MR system (K012200). The submission focuses on demonstrating substantial equivalence rather than a new device with novel performance claims requiring extensive clinical outcome acceptance criteria.

Therefore, the "acceptance criteria" for this device are primarily related to performance and safety standards adherence and substantial equivalence to the predicate device, rather than diagnostic accuracy or effectiveness against a specific disease state. The "study" described is a technical evaluation rather than a clinical trial.

Here's a breakdown of the requested information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Sample Size: The document does not describe a "test set" in the context of clinical images or patient data for evaluating diagnostic performance. The evaluation focused on device performance against technical and safety standards and comparison to a predicate device.
• Data Provenance: Not applicable, as there was no clinical test set described. The "data" would be engineering and performance measurements.

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

• Number of Experts: Not applicable. Ground truth typically refers to clinical diagnosis or pathology for a medical image analysis device. This submission is for an MRI system itself, not an AI or image analysis tool.
• Qualifications: Not applicable.

4. Adjudication Method (e.g., 2+1, 3+1, none) for the Test Set

• Adjudication Method: Not applicable. No clinical test set requiring expert adjudication was described.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size of How Much Human Readers Improve with AI vs. Without AI Assistance

• MRMC Study: No, an MRMC comparative effectiveness study was not done. This device is the MRI scanner itself, not an AI assistance tool for human readers.
• Effect Size: Not applicable.

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

• Standalone Performance: Not applicable. This document describes an MRI system, which produces images for interpretation by a human physician, not a standalone algorithm for diagnosis.

7. The Type of Ground Truth Used (expert consensus, pathology, outcomes data, etc.)

• Type of Ground Truth: For the "study" described (evaluation against standards and comparison to predicate), the "ground truth" would be the established NEMA, IEC 601-1, and IEC 601-2-33 standards for MRI system performance and safety, and the performance characteristics of the predicate Signa TwinSpeed MR system. There isn't a clinical "ground truth" in the sense of patient outcomes or pathology.

8. The Sample Size for the Training Set

• Sample Size: Not applicable. This MRI system is not an AI model that requires a "training set" of patient data in the computational sense. The development of the system itself would involve extensive engineering and testing, but these are not referred to as "training sets" in this context.

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

• How Ground Truth Was Established: Not applicable. As explained above, there is no "training set" for this type of medical device submission.

Ask a specific question about this device