Search Results

AIR Recon DL is a deep learning based reconstruction technique that is available for use on GE HealthCare 1.5T, 3.0T, and 7.0T MR systems. AIR Recon DL reduces noise and ringing (truncation artifacts) in MR images, which can be used to reduce scan time and improve image quality. AIR Recon DL is intended for use with all anatomies, and for patients of all ages. Depending on the anatomy of interest being imaged, contrast agents may be used.

AIR Recon DL is a software feature intended for use with GE HealthCare MR systems. It is a deep learning-based reconstruction technique that removes noise and ringing (truncation) artifacts from MR images. AIR Recon DL is an optional feature that is integrated into the MR system software and activated through purchasable software option keys. AIR Recon DL has been previously cleared for use with 2D Cartesian, 3D Cartesian, and PROPELLER imaging sequences.

The proposed device is a modified version of AIR Recon DL that includes a new deep-learning phase correction algorithm for applications that create multiple intermediate images and combine them, such as Diffusion Weighted Imaging where multiple NEX images are collected and combined. This enhancement is an optional feature that is integrated into the MR system software and activated through an additional purchasable software option key (separate from the software option keys of the predicate device).

This document describes the acceptance criteria and the studies conducted to prove the performance of the AIR Recon DL device, as presented in the FDA 510(k) clearance letter.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Nonclinical Testing:
  • Phantom testing was conducted for the DLPC and PC-ARDL models. No specific sample size (number of phantom scans) is provided, but it implies a sufficient number for evaluation.
• In-Vivo Performance Testing:
  • ADC Accuracy: Diffusion-weighted brain images were acquired at 1.5T with b-values = 50, 400, 800, 1200 s/mm². The number of subjects is not explicitly stated, but it's referred to as "diffusion images" and "diffusion-weighted brain images."
  • Low-Contrast Detectability: Raw data from 4 diffusion-weighted brain scans were used.
• Quantitative Post Processing (Repeatability Study):
  • 6 volunteers were recruited. 2 volunteers scanned on a 1.5T scanner, 4 on a 3T scanner.
  • Scanned anatomical regions included brain, spine, abdomen, pelvis, and breast.
  • Each sequence was repeated 4 times.
  • Data Provenance: The document states "in-vivo data" and "volunteer scanning was performed simulating routine clinical workflows." This suggests prospective scanning of human subjects, likely in a controlled environment. The country of origin is not specified, but given the FDA submission, it's likely U.S. or international data meeting U.S. standards. The statement "previously acquired de-identified cases" for the Clinical Image Quality Study refers to retrospective data for that specific study, but the volunteer scanning for repeatability appears prospective.
• Clinical Image Quality Study:
  • 34 datasets of previously acquired de-identified cases.
  • Data Provenance: "previously acquired de-identified cases" indicates retrospective data. The country of origin is not specified.

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

• Nonclinical Testing: Ground truth established through phantom measurements and expected physical properties (e.g., signal bias, noise floor). No human experts involved in establishing ground truth here.
• In-Vivo Performance Testing:
  • ADC Accuracy: "Average ADC values were measured from regions of interest in the lateral ventricles." This implies expert selection of ROIs, but the number of experts is not specified. The ground truth for ADC is the expected isotropic Gaussian diffusion in these regions.
  • Low-Contrast Detectability: "The contrast ratio and contrast-to-noise ratio for each of the inserts were measured." This is a quantitative measure, not explicitly relying on expert consensus for ground truth on detectability, but rather on the known properties of the inserted synthetic objects.
• Quantitative Post Processing:
  • ADC Repeatability: Ground truth for repeatability is based on quantitative measurements and statistical analysis (coefficient of variability). ROI placement would typically be done by an expert, but the number is not specified.
  • Phase Correction Effectiveness: Ground truth is based on the theoretical expectation of signal distribution in real/imaginary channels after ideal phase correction.
• Clinical Image Quality Study:
  • One (1) U.S. Board Certified Radiologist was used.
  • Qualifications: "U.S. Board Certified Radiologist." No explicit number of years of experience is stated, but Board Certification indicates a high level of expertise.

4. Adjudication Method for the Test Set

• Nonclinical/Phantom Testing: No explicit adjudication method described beyond passing defined acceptance criteria for quantitative metrics.
• In-Vivo Performance Testing: Quantitative measurements (ADC values, contrast ratios, CNR) were used. Paired t-tests were conducted, which is a statistical comparison method, not an adjudication process as typically defined for expert readings.
• Quantitative Post Processing: Quantitative measurements and statistical analysis (coefficient of variability, comparison of real/imaginary channels).
• Clinical Image Quality Study: A single U.S. Board Certified Radiologist made the assessment. There is no stated adjudication method described, implying a single-reader assessment for clinical image quality.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• An MRMC comparative effectiveness study was not explicitly described as a formal study design in the provided text.
• The "Clinical Image Quality Study" involved only one radiologist, so it does not qualify as an MRMC study.
• There is no reported effect size of how much human readers improve with AI vs. without AI assistance. The study rather focused on the AI-reconstructed images' standalone diagnostic quality.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

• Yes, performance was evaluated in a standalone manner.
  • Nonclinical Testing: Phantom studies directly evaluate the algorithm's output against known physical properties and defined metrics.
  • In-Vivo Performance Testing: ADC accuracy and low-contrast detectability were measured directly from the reconstructed images, which is a standalone evaluation of the algorithm's quantitative output.
  • Quantitative Post Processing: Repeatability and effectiveness of phase correction in real/imaginary channels are algorithm-centric evaluations.
  • Even the clinical image quality study, while involving a human reader, assessed the standalone output of the algorithm (AIR Recon DL with Phase Correction) for diagnostic quality.

7. Type of Ground Truth Used

• Expert Consensus: Not explicitly stated as the primary ground truth for quantitative metrics, but one radiologist's assessment served as the primary clinical ground truth.
• Pathology: Not used as ground truth in the provided study descriptions. While some datasets "included pathological features such as prostate cancer... hepatocellular carcinoma," the assessment by the radiologist was on "diagnostic quality" of the images, not a comparison against pathology reports for definitive disease identification.
• Outcomes Data: Not used as ground truth.
• Other:
  • Physical Properties/Known Standards: For phantom testing (e.g., signal bias, noise floor, SNR, sharpness), and for theoretical expectations of ADC values in specific regions (lateral ventricles).
  • Known Synthetic Inserts: For low-contrast detectability.
  • Theoretical Expectations: For phase correction effectiveness (signal in real, noise in imaginary).

8. Sample Size for the Training Set

• The document does not provide any specific sample size for the training set used for the deep learning models (DLPC and PC-ARDL). It only states that the models are "deep learning-based."

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

• The document does not provide any information on how the ground truth for the training set was established. It only describes the testing of the final, trained models.

Ask a specific question about this device

AIR Recon DL is a deep learning based reconstruction technique that is available for use on GE Healthcare 1.5T, 3.0T, and 7.0T MR systems. AIR Recon DL reduces noise and ringing (truncation artifacts) in MR images, which can be used to reduce scan time and improve image quality. AIR Recon DL is intended for use with all anatomies, and for patients of all ages. Depending on the anatomy of interest being imaged, contrast agents may be used.

AIR Recon DL is a software feature intended for use with GE Healthcare MR systems. It is a deep learning based reconstruction technique that removes noise and ringing (truncation) artifacts from MR images. AIR Recon DL is an optional feature that is integrated into the MR system software and activated through a purchasable software option key.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary for AIR Recon DL:

1. Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the desired improvements and non-inferiority claims of the AIR Recon DL feature. The reported device performance demonstrates how these criteria were met.

2. Sample Size and Data Provenance

• Test Set Sample Size: 133 cases.
  • 129 patient cases
  • 4 healthy subjects
• Data Provenance:
  • Country of Origin: Not explicitly stated, but "10 different clinical sites" suggests a multi-center study, and "a GE Healthcare facility" could indicate a US or international site. Given the FDA submission, it's highly likely to include US data.
  • Retrospective or Prospective: Not explicitly stated, but the description "images acquired across a variety of pulse sequences and anatomies" and involvement of "10 different clinical sites" could imply a retrospective collection for the reader study, where images were pre-collected. However, without explicit mention, it's not definitive. The phrasing "acquired from the same acquired raw data" suggests a paired comparison based on existing data.

3. Number of Experts and Qualifications

• Number of Experts: Three radiologists.
• Qualifications: "Radiologists" implies medical doctors specialized in radiology. No further specifics on years of experience or subspecialty were provided.

4. Adjudication Method for the Test Set

The adjudication method appears to be 2+1 (or 3/3 agreement is ideal, but 2 out of 3 for consensus is common).
"Each image pair was evaluated independently by three radiologists."
"The results confirmed that the AIR Recon DL feature provides images with equivalent or better image quality in terms of apparent signal to noise ratio (133 out of 133 cases), sharpness (133 out of 133 cases), and lesion conspicuity (123 out of 124 cases with pathology)."
"The radiologists reading the images also indicated a preference for the AIR Recon DL images over conventional images in 99% of the evaluations."
This indicates that the claims are based on the collective agreement or majority opinion of the three readers for each case. The exact decision rule (e.g., simple majority, unanimous) is not stated, but the high consistency (e.g., 133/133, 123/124) implies strong agreement or effective resolution.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

Yes, a multi-reader multi-case (MRMC) comparative effectiveness study was done.

• Comparison: Radiologists compared AIR Recon DL images to conventional images (without AIR Recon DL) reconstructed from the same acquired raw data.
• Effect Size of Human Reader Improvement: The study demonstrates a significant preference and improvement in perceived image quality attributes by human readers when using AIR Recon DL assisted images compared to conventional images.
  • Preference: Radiologists preferred AIR Recon DL images over conventional images in 99% of evaluations. This indicates a strong positive effect on reader perception and diagnostic confidence.
  • Qualitative Improvement:
    • Equivalent or better SNR in 100% of cases (133/133).
    • Equivalent or better sharpness in 100% of cases (133/133).
    • Equivalent or better lesion conspicuity in ~99.2% of pathological cases (123/124).
  • Enablement of Shorter Scans: For shorter scan time acquisitions, AIR Recon DL images were rated as better or equivalent image quality in 100% of 22 image pairs, which suggests human readers are able to maintain or even improve their assessment quality despite reduced acquisition time. Overall, the effect size is very large and consistently positive across all measured subjective criteria.

6. Standalone (Algorithm Only) Performance

The document describes "nonclinical testing" on phantoms, which represents a form of standalone testing where the algorithm's output is directly measured against predefined physical criteria:

• "AIR Recon DL has undergone phantom testing to evaluate the feature and its impact on image quality, including SNR, sharpness, and low contrast detectability."
• "The nonclinical testing demonstrated that AIR Recon DL does improve SNR and image sharpness while maintaining low contrast detectability."
• "ADC maps were not adversely impacted by the use of AIR Recon DL."
  This evaluates the algorithm's effect on image characteristics absent human interpretation of clinical cases.

7. Type of Ground Truth Used

The study primarily used expert consensus (radiologist agreement) as the ground truth for evaluating image quality attributes (SNR, sharpness, lesion conspicuity, overall preference) and the impact of the algorithm.
For the "presence of pathology" in the test set, it's assumed that this was either identified beforehand from clinical reports or pathology (e.g., biopsy results) or by consensus among the evaluating radiologists prior to their evaluation of the AI-enhanced images. However, the exact source of ground truth for pathology presence/absence isn't detailed, only that 124 cases "with pathology" and 9 cases "without pathology" were included.

8. Sample Size for the Training Set

The document states: "Both the proposed AIR Recon DL and the predicate device use neural networks that have similar architecture, and were trained using similar methods and data."
However, the specific sample size for the training set is NOT provided in this summary.

9. How Ground Truth for Training Set was Established

The document states: "Both the proposed AIR Recon DL and the predicate device use neural networks that have similar architecture, and were trained using similar methods and data."
The method for establishing ground truth for the training set is NOT explicitly detailed. Typically, for deep learning reconstruction, the "ground truth" during training often refers to high-quality, fully sampled MR images (or images from a prior, high-quality reconstruction method) that the AI attempts to match or improve upon, rather than a clinical diagnosis per se. The goal during training would be to generate images that are less noisy and sharper while preserving underlying anatomical and pathological information, learned by comparing "corrupted" (e.g., undersampled, noisy) inputs to "clean" reference images.

Ask a specific question about this device