Search Results

The Deep Learning Image Reconstruction software is a deep learning based reconstruction method intended to produce cross-sectional images of the head and whole body by computer reconstruction of X-ray transmission data taken at different angles and planes, including Axial, (Volumetric), and Cardiac acquisitions, for all ages.

Deep Learning Image Reconstruction software can be used for head, whole body, cardiac, and vascular CT applications.

Deep Learning Image Reconstruction is an image reconstruction method that uses a dedicated Deep Neural Network (DNN) that has been designed and trained specifically to generate CT Images to give an image appearance, as shown on axial NPS plots, similar to traditional FBP images while maintaining the performance of ASiR-V in the following areas: image noise (pixel standard deviation), low contrast detectability, high-contrast spatial resolution, and streak artifact suppression.

The images produced are branded as "TrueFidelity™ CT Images". Reconstruction times with Deep Learning Image Reconstruction support a normal throughput for routine CT.

The deep learning technology is integrated into the scanner's existing raw data-based image reconstruction chain to produce DICOM compatible "TrueFidelity™ CT Images".

The system allows user selection of three strengths of Deep Learning Image Recon: Low, Medium or High. The strength selection will vary with individual users' preference for the specific clinical need.

The DLR algorithm was modified on the Revolution CT/Apex platform for improved reconstruction speed and image quality and cleared in February 2022 with K213999. The same modified DLIR is now being ported to Revolution EVO (K131576) /Revolution Maxima (K192686), Revolution Ascend (K203169, K213938) and Discovery CT750 HD family CT systems including Discovery CT750 HD, Revolution Frontier and Revolution Discovery CT (K120833).

The provided text describes that the Deep Learning Image Reconstruction software was tested for substantial equivalence to a predicate device (K213999). The study performed was largely an engineering bench testing, comparing various image quality metrics between images reconstructed with Deep Learning Image Reconstruction (DLIR) and ASiR-V from the same raw datasets.

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

1. Table of Acceptance Criteria and Reported Device Performance

The text indicates that the device aims to maintain the performance of ASiR-V in specific areas while offering an image appearance similar to traditional FBP images. The "acceptance criteria" can be inferred from the list of image quality metrics evaluated, with the performance goal being comparable or improved relative to ASiR-V.

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

• Test Set Sample Size: The text states "the identical raw datasets obtained on GEHC's Revolution Ascend, Revolution Frontier and Discovery CT750 HD systems". However, the number of cases or specific sample size for these datasets is not explicitly stated.
• Data Provenance: The raw datasets were "obtained on GEHC's Revolution Ascend, Revolution Frontier and Discovery CT750 HD systems". The country of origin is not specified, and it is stated that the study used retrospective raw datasets (i.e., existing data, not newly acquired data for the study).

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

The provided text focuses on engineering bench testing and image quality metrics. It does not mention the use of experts to establish ground truth for the test set or their qualifications. The evaluation primarily relies on quantitative image quality metrics.

4. Adjudication Method for the Test Set

Since experts were not explicitly used to establish ground truth, there is no mention of an adjudication method for the test set in the provided text.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done, and its effect size

An MRMC comparative effectiveness study was not performed according to the provided text. The study focused on technical image quality comparisons at the algorithm level, not human reader performance with or without AI assistance.

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

Yes, a standalone performance evaluation was done. The study described is primarily a standalone evaluation of the algorithm's image quality output (e.g., noise, resolution, artifacts, detectability) when compared to images reconstructed with ASiR-V from the same raw data.

7. The Type of Ground Truth Used

The "ground truth" for the test set was essentially:

• Quantitative Image Quality Metrics: Performance relative to ASiR-V for metrics like image noise, LCD, spatial resolution, streak artifact suppression, CT uniformity, CT number accuracy, CNR, spatial resolution (longitudinal), NPS, and artifact analysis.
• Reference Image Appearance: The stated goal was an image appearance similar to traditional FBP images shown on axial NPS plots.

There is no mention of pathology, expert consensus on clinical findings, or outcomes data being used as ground truth for this particular substantial equivalence study.

8. The Sample Size for the Training Set

The text states that the Deep Neural Network (DNN) is "trained on the CT scanner" and models the scanned object using "information obtained from extensive phantom and clinical data." However, the specific sample size for the training set is not provided.

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

The ground truth for the training set is implicitly established through the "extensive phantom and clinical data" mentioned as being used to train the DNN. The text indicates the DNN is trained to model noise propagation and identify noise characteristics to remove it, and to generate images with an appearance similar to traditional FBP while maintaining ASiR-V performance. This suggests the training involves learning from "ground truth" as defined by:

• Reference Image Quality: Likely images reconstructed with proven methods (e.g., FBP, ASiR-V) or images from phantoms with known properties.
• Noise Characteristics: The DNN is trained to understand and model noise.

However, the specific methodology for establishing this ground truth for the training data (e.g., expert annotation, simulated data, pathology confirmed disease) is not detailed in the provided text.

Ask a specific question about this device

The system is intended to produce cross-sectional images of the body by computer reconstruction of x-ray transmission data taken at different angles and planes, including Axial, Cine, Helical (Volumetric), Cardiac, and Gated acquisitions. These images may be obtained either with or without contrast. This device may include signal analysis and display equipment, patient and equipment supports, components and accessories.

This device may include data and image processing to produce images in a variety of trans-axial and reformatted planes. Further the images can be post processed to produce additional imaging planes or analysis results.

The system is indicated for head, whole body, cardiac and vascular X-ray Computed Tomography applications in patients of all ages.

The device output is a valuable medical tool for the diagnosis of disease, trauma, or abnormality and for planning, guiding, and monitoring therapy.

The Revolution Ascend is a head and whole-body CT system composed of a gantry, patient table, operator console with a host computer, power distribution unit, and interconnecting cables. The system also includes image acquisition and reconstruction hardware/software, general system software, accompanying documents, and associated accessories/interconnections. The system has a 75 cm gantry bore and 64-row detector.

Revolution Ascend generates cross-sectional images of the body by computer reconstruction of x-ray transmission data taken at different angles and planes, including Axial, Cine, Helical (Volumetric), Cardiac, and Gated acquisitions modes.

A design change has been made to the Revolution Ascend with an alternative detector scintillator material prompting this premarket notification. While this change is being made, the design and manufacturing is such that the system performance remains identical to its unmodified predicate. The proposed device carries over all the features, options and specifications of the predicate device, including the Deep Learning Iterative Recon (DLIR) cleared via K212067 without change.

This document is a 510(k) Premarket Notification Summary for the Revolution Ascend CT system. The purpose of this submission is to demonstrate that the proposed device, with a change in detector scintillator material, is substantially equivalent to a legally marketed predicate device. Therefore, the acceptance criteria and study design are focused on proving this equivalence rather than establishing the de novo performance of an AI algorithm or a new medical device.

Based on the provided document, here's a description of the acceptance criteria and the study that proves the device meets them:

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

The document doesn't provide a direct table of specific numerical acceptance criteria for image quality metrics. Instead, the acceptance criteria are implicitly stated as demonstrating equivalence to the predicate device, Revolution Ascend (K203169), across various performance aspects.

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

The document explicitly states that the testing was non-clinical bench testing using "standard IQ, QA phantoms." It does not involve human patient data or a specific "test set" in the context of clinical studies. Therefore, sample size in terms of patient cases is not applicable here.

• Data Provenance: Not applicable as it's non-clinical bench testing with phantoms.

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

Not applicable. As the testing was non-clinical bench testing using phantoms and established metrics (e.g., IEC standards, NEMA standards), the "ground truth" is based on the known physical properties and performance characteristics of the phantoms and the objective measurements derived from them, rather than expert interpretation of patient images.

4. Adjudication method for the test set

Not applicable. Since the testing is non-clinical bench testing with phantoms and objective measurements, there is no need for expert adjudication of image findings.

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. This submission is for a hardware change (detector scintillator material) in a CT system, not for a new AI-powered diagnostic device or a modification to an existing AI feature (DLIR is carried over without change). Therefore, an MRMC comparative effectiveness study regarding human reader performance with/without AI assistance is outside the scope of this particular 510(k) submission. The document explicitly states the DLIR was "cleared via K212067 without change," implying its performance was evaluated in that separate submission.

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

No. This submission is for a CT scanner system that includes hardware and software. It's not for a standalone algorithm. The "Deep Learning Image Reconstruction (DLIR)" component referenced is a reconstruction algorithm within the CT system, and its standalone performance likely would have been assessed in its original 510(k) clearance (K212067). This submission focuses on demonstrating that the change in detector material does not degrade the performance of the overall system, including features like DLIR.

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

For the non-clinical testing, the "ground truth" is based on objective phantom measurements and established engineering metrics as defined by standards like IEC 61223-3-5 Ed. 2. This is not clinical ground truth (e.g., pathology, expert consensus on disease diagnosis). The goal is to demonstrate physical and image quality equivalence.

8. The sample size for the training set

Not applicable. This submission is about a hardware change in an already cleared CT system and is not for training a new AI algorithm. The DLIR component, which involves deep learning, would have had a training set in its original development and clearance (K212067), but details for that are not provided in this document as it's "carried over without change."

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

Not applicable. As above, this pertains to the development of the DLIR algorithm (likely cleared in K212067), not the current submission for a detector material change.

Ask a specific question about this device