Technology | Computed Tomography (CT) | April 19, 2019

FDA Clears GE's Deep Learning Image Reconstruction Engine

Engine provides TrueFidelity CT images for Revolution, Revolution Apex CT systems; FDA also clears trio of CT applications

FDA Clears GE's Deep Learning Image Reconstruction Engine

April 19, 2019 – GE Healthcare has received 510(k) clearance from the U.S. Food and Drug Administration (FDA) of its Deep Learning Image Reconstruction engine on the new Revolution Apex computed tomography (CT) device. The engine has also been approved as an upgrade to GE’s Revolution CT system in the United States.

Deep Learning Image Reconstruction (DLIR) is the next-generation image reconstruction option that uses a dedicated deep neural network (DNN) to generate TrueFidelity CT Images. These TrueFidelity CT Images have the potential to improve reading confidence in a wide range of clinical applications such as head, whole body and cardiovascular, for patients of all ages.

Compared to current iterative reconstruction technology, TrueFidelity CT Images can elevate every image to a powerful first impression with impressive image quality performance[1], and preferred image sharpness and noise texture[2], without compromising dose performance, according to GE.

The company noted that the Revolution Apex CT includes the Quantix 160 X-ray tube, which is also an optional upgrade for the Revolution CT.

The FDA also granted 510(k) clearance to three additional GE Healthcare CT applications introduced at the 2018 Radiological Society of North America annual meeting (RSNA 2018): Bone VCAR, Thoracic VCAR with GSI Pulmonary Perfusion and SnapShot Freeze 2.

Both Bone VCAR and SnapShot Freeze 2 are Edison applications. Edison is a platform that helps accelerate the development and adoption of artificial intelligence (AI) technology and empower providers to deliver faster, more precise care.

Bone VCAR uses a deep-learning algorithm to automatically identify and label the vertebrae for faster spine assessment and improve reporting efficiency. SnapShot Freeze 2 is an intelligent motion correction algorithm that expands the breadth of intelligent motion correction applications, further enhancing clinical performance and introducing whole heart motion correction.

Thoracic VCAR with GSI Pulmonary Perfusion provides automated segmentation and measurements for assessment and follow-up of thoracic diseases. It also now includes dedicated GSI Pulmonary Perfusion, a one-click protocol to assist in the identification of variations of relative perfusion distribution in the lungs. Thoracic diseases that may be associated with changes in perfusion include pulmonary embolism or chronic obstructive pulmonary disease (COPD).

For more information: www.gehealthcare.com

 

References

[1] Image quality comparisons between DLIR and ASiR-V, were evaluated by phantom tests of MTF, SSP, axial NPS, standard deviation of image noise, CT Number accuracy, CNR, and artefact analysis.  Additionally, LCD was demonstrated in phantom testing using a model observer with the head and body MITA CT IQ Phantoms (CT191, CT189 The Phantom Laboratory). DLIR and ASiR‐V reconstructions were performed using the same raw data.

[2] As demonstrated in a clinical evaluation consisting of 60 cases and 9 physicians, where each case was reconstructed with both DLIR and ASiR‐V and evaluated by 3 of the physicians. In 100% of the reads, DLIR’s image sharpness was rated the same as or better than ASiR‐V’s.  In 91% of the reads, DLIR’s noise texture was rated better than ASiR‐V’s. This rating was based on each individual reader’s preference.

Related Content

Philips launches Azurion with FlexArm to set new standard for the future of image-guided procedures

Philips launches Azurion with FlexArm to set new standard for the future of image-guided procedures.

News | Radiology Imaging | December 12, 2019
December 12, 2019 — Philips and the Regional Medical Center
Schematic diagram of the proposed multichannel deep neural network model analyzing multiscale functional brain connectome for a classification task. rsfMRI = resting-state functional MRI.

Schematic diagram of the proposed multichannel deep neural network model analyzing multiscale functional brain connectome for a classification task. rsfMRI = resting-state functional MRI. Graphic courtesy of the Radiological Society of North America.

News | Artificial Intelligence | December 11, 2019
December 11, 2019 — Deep learning, a type of arti...
CT_Pediatric_Scan_Philips_Vereos_CT_RSNA 2016

Image courtesy of Philips Healthcare

News | Pediatric Imaging | December 10, 2019
December 10, 2019 — More than half of people who received...
Damage from concussion alters the way information is transmitted between the two halves of the brain, according to a new study presented today at the annual meeting of the Radiological Society of North America (RSNA).

Image courtesy of RSNA

News | Clinical Trials | December 10, 2019
December 10, 2019 — Damage from...
 MRI system cardiac scan from DeBakey Hospital
News | Magnetic Resonance Imaging (MRI) | December 05, 2019
December 5, 2019 — The following is by Lawr