Source: Equipment Home, reprinting is not allowed without authorization, and it can only be reproduced 24 hours later.
Recently, Equipment Home learned that GE Medical announced that the US FDA 510(k) has approved its breakthrough AIR Recon DL for 3D and propeller imaging sequences. These new features extend the benefits of AIR Recon DL to nearly all magnetic resonance imaging (MRI) clinical programs covering all anatomical structures for better image quality, shorter scanning times and a better patient experience.
GE Medical's AIR Recon DL based on deep learning MRI image reconstruction technology has fundamentally changed the balance between image quality and scanning time. Now, with the solution’s extended compatibility from 2D to 3D imaging sequences, doctors can quickly and confidently diagnose patients, improving signal-to-noise ratio (SNR) and clarity. 3D imaging provides higher clinical efficiency, allowing radiologists to no longer need multiple 2D acquisitions, potentially speeding up diagnosis.
01
AIR What kind of technology is Recon DL?
AIR™ Recon DL is the first deep learning reconstruction technology in the industry to be approved by GE Healthcare. In AIR™ Recon DL, after tens of millions of fitting iterations, the deep-optimized deep- convolutional neural network model subverts the traditional reconstruction concept. This technology embeds deep learning technology into the original data stage of magnetic resonance reconstruction for the first time, which realizes the effective separation of MR signals and noise signals, thereby obtaining pure magnetic resonance signals, greatly improving the efficiency of magnetic resonance imaging, making the selection of scanning parameters no longer a major factor restricting image quality, and broadening the new boundaries between magnetic resonance scanning speed, resolution and signal-to-noise ratio.
AIR Recon DL was developed in collaboration with institutions around the world and has been evaluated in thousands of cases of various anatomical structures and patient demographics. Feedback from clinical users is very positive, including users noting clearer and less noise images, which reduces scanning time, increases confidence, reduces the need for repeated scans, and improves consistency between scans.
AIR Recon DL's development and clinical validation partners include: Special Surgery Hospital, University of California, San Francisco, RadNet, University of Wisconsin-Madison, MD Anderson Cancer Center, Wisconsin Medical College , French Northern Heart Disease Center, Netherlands Erasmus Medical Center, Italy Centro Cardiologico, Yamanashi University and Keio University in Japan, a San Medical Center in Korea and Haeundae Paik Hospital.
AIR Recon DL was developed on GE H Medical's Edison Intelligence Platform, which helps GE and strategic partners quickly design, develop, manage, protect, and distribute advanced applications and AI algorithms.
AIR™ Recon DL AIR™ Recon DL AIR™ In 2019, GE Healthcare introduced the magnetic resonance imaging reconstruction process in the industry, and obtained FDA approval to allow use on clinical 3.0T. In AIR™ Recon DL, the defined multi-layer deep convolutional neural network has been calculated through tens of millions of times of fitting iteration to obtain a set of millions of weighting factors. Based on the optimized deep convolutional neural network model, the data collected by magnetic resonance imaging is different from the traditional reconstruction method. The effective information of will be automatically extracted for reconstruction. The obtained image has the following characteristics:
, significantly reducing noise level, thereby improving the image signal-to-noise ratio. Note: The choice of noise reduction level can be based on the user's specific preferences, including three options: 25%, 50% and 75%,
, significantly improved image resolution and sharper inter-tissue contrast;
, significantly reduced image artifacts caused by motion and other reasons.
The increased image signal-to-noise ratio and improved image contrast make the setting of scanning parameters no longer demanding during the scanning process, so that the scanning speed can be increased by 30%-50% , thus better serving the clinical service. Combining these advantages, this technology has conducted in-depth exploration on the structural and functional imaging of various parts of the body including nerves, breast , body, bone muscles, and other parts of the body, and has achieved fruitful scientific research results.
02
Take Nervous system as an example
Solve the last mile clinical problem
second instead of minutes to become a new unit of measurement for scientific research scanning. Hundred-second imaging usually includes submillimeter voxel structure imaging, high-resolution functional imaging, and quantitative map imaging. Neurological imaging is the most concerned direction in magnetic resonance imaging research. With the underlying reconstruction of AIR™ Recon DL, has constructed a new balance of scanning time, image resolution, and image signal-to-noise ratio, which has also opened a new chapter for neural imaging.
Ultra-fast imaging
Ultrafast imaging has always been an important direction for the development of magnetic resonance technology. In clinical diagnostic scans, rapid imaging means higher diagnostic efficiency, providing timely imaging for emergency diagnosis, thereby meeting the increasing clinical diagnosis needs. However, the improvement of imaging speed alone often leads to a decrease in imaging quality, including a decrease in signal-to-noise ratio, a decrease in contrast and an increase in image artifacts. This is a challenge for the diagnosis of neurological diseases that require especially high-definition imaging as the basis for diagnosis. So, can fast and high-quality imaging be obtained simultaneously in neuromagnetic resonance scans?
AIR™ Recon DL technology gives the answer: is based on three major features of AIR™ Recon DL: high signal-to-noise ratio, high image resolution and sharp contrast, and more realistic image restoration after removing artifacts. It is no longer harsh in the strict scanning parameter settings required for high-quality imaging when conventional clinical head magnetic resonance scanning sequences are no longer demanding. uses AIR™ Recon DL reconstruction technology, and the clinical head magnetic resonance scanning protocol requires less than 1/4 of the original scanning time (traditional 10-minute imaging, AIR™ Recon DL only takes 2 minutes), and can obtain imaging quality comparable to traditional reconstructed images, thus achieving ultra-fast and high-quality imaging in the true sense.
HD tissue imaging
Brain structure imaging with high signal-to-noise ratio and strong tissue contrast is the necessary imaging basis for studying neurological diseases including degenerative diseases, epilepsy, Multiple sclerosis, Multiple sclerosis, and other neurological diseases. T2w-PROPELLER imaging is widely recognized in the clinical diagnosis of intracranial nerves because it can effectively inhibit the artifacts caused by human movement and fluid flow. However, the long scanning time required for high-resolution T2w-PROPELLER imaging to obtain sufficient image signal-to-noise ratio and tissue contrast make it challenging to be widely used in clinical practice.
combined with AIR™ Recon DL reconstruction technology, the image resolution reaches sub-mm-level ultra-high-definition T2w-PROPELLER (0.2 x 0.2 x 1mm3, PROPELLER-DL) brain imaging. Compared with the images obtained by traditional reconstruction methods (PROPELLER), the image signal-to-noise ratio has been significantly improved. More importantly, the microstructure of brain functional areas such as the amygdala on PROPELLER-DL, hippocampus and cerebellar (deep nucleus, cerebellar cortex and vermis) can be clearly displayed.
Neurovascular imaging
Neurological and vascular tissue are core components of the human brain tissue structure, and the evaluation of its morphology and function is an important basis for the diagnosis of clinical neurologic diseases. Due to its tiny structure, the resolution of neural and vascular magnetic resonance imaging often needs to reach the order of millimeters or even submillimeters. therefore requires a longer scanning time, resulting in (motion) artifacts, lower image signal-to-noise ratio and insufficient tissue contrast limit its wide application in clinical practice.
combined with AIR™ Recon DL reconstruction technology, 2D DIXON and 3D CUBE-STIR imaging have even been shortened. The obtained image signal-to-noise ratio of brachial plexus nerve root imaging (DIXON-DL, CUBE-STIR-DL), the contrast between nerve roots and surrounding tissues, and the display of local nerve details (including: nerve roots, nerve stems, nerve bifurcations, etc.) have been significantly improved, providing a solid theoretical basis for its application in neuroimaging.
Small vessel imaging
At the same time, in terms of small vessel imaging, T2* or SWI imaging based on 3D EPI can complete isotropic vascular imaging with ultra-high image resolution (0.5mm3) in less than 4 minutes of scanning time. However, the lower image signal-to-noise ratio makes it challenging in the detection and diagnosis of small blood vessel-related diseases (such as microbleeding, calcification or iron deposition).
However, with the help of AIR™ Recon DL technology, the reconstructed T2* (T2*-DL) and SWI (SWI-DL) angiogenesis have significantly improved compared with the T2* and SWI imaging obtained by traditional reconstruction methods in image signal-to-noise ratio, overall image quality, image contrast and display of tiny blood vessels. In addition to , T2*-DL and SWI-DL after two repetitive scans have higher consistency in the display of image details.
Accurate quantitative map
MAGiC Due to its one-stop infinite contrast structure imaging and comprehensive relaxation quantitative map, its value in clinical neuroscience research has been widely recognized. further combines AIR™ Recon DL technology to significantly improve the original data used for image synthesis in signal-to-noise ratio, contrast, and real data restoration after artifact removal.
hardcover function evaluation
diffusion magnetic resonance imaging, because it can truly reflect the diffusion characteristics of tissue molecules, brain network connections and microstructure, it is widely used in clinical diagnosis, especially in the field of clinical nerves. However, due to the limitations of the imaging principle, the collected diffuse imaging often cannot have the characteristics of high signal-to-noise ratio, high tissue contrast, few imaging artifacts, small image deformation and short scanning time.
Among them, PROPELLER diffusion imaging (DW-PROPELLER) has the characteristics of few artifacts and small deformation, and is suitable for evaluating relatively uneven local magnetic fields such as the skull base and head and neck. However, obtaining sufficient signal-to-noise ratio and tissue contrast imaging in a short scanning time is a bottleneck that hinders its further clinical widespread application.
combined with AIR™ Recon DL technology without increasing imaging time, the signal-to-noise ratio and tissue contrast of the reconstructed PROPELLER diffusion (DW- PROPELLER-DL) are significantly improved compared with traditional reconstructed images, reaching a level comparable to that of multiple excitation diffusion imaging (MUSE), and are better than MUSE imaging in artifact suppression and image deformation control. What is more worth mentioning is that the quantitative parameter ADC derived from DW- PROPELLER-DL shows smaller value fluctuations in model experiments and more realistically reduces its molecular diffusion characteristics.
03
Improve MR imaging Experience
can be used for 1.5T, 3.0T Most models upgrade
AIR Recon DL is now compatible with PROPELLER, an imaging sequence that is insensitive to movement and is particularly important for anatomical structures that are susceptible to movement (such as breathing) during MRI examinations, as well as pediatric, neurodegenerative, elderly and claustrophobic patients who are difficult to keep their bodies still during MRI scans. Therefore, doctors can obtain clearer images without repeated scans, thereby increasing patient processing volume, improving schedule, shortening examination time and speeding up diagnosis, helping to create a full range of better patient experiences.The original intention of the research and development of
AIR™ Recon DL technology is to challenge all the most difficult pain points in clinical practice, and solve the problem of the last mile of clinical practice. As of September 2022, at least 3.5 million patients worldwide have undergone AIR Recon DL scans. According to recent AIR Recon DL 3D and PROPELLER reader studies:
00% of participants said that deep learning solutions provide better signal-to-noise ratios and better or comparable image clarity than conventional image reconstruction;
99% also said that AIR Recon DL provides better or considerable damage significance;
report also showed that exam time was reduced by 50%, improving workflow efficiency and productivity.
"By putting AIR Recon DL expands to 3D and PROPELLER, GE Healthcare narrows the gap in our ability to provide better image quality and patient experience to patients of all examination types, especially in brain imaging and musculoskeletal imaging that we rely heavily on 3D sequences. PROPELLER is important for reducing image quality variability and eliminating repetitions caused by movement," said Tiron Pechet, radiologist and assistant medical director at Shields Health Care Group.
AIR Recon DL and recent expansions can be used for the new GE Medical MRI scanner and as an upgrade to most 1.5T and 3.0T MRI systems installed by GE Medical, in addition to its efficiency and productivity benefits to the industry, this upgrade will enable the previously installed scanner to be updated and run like a new system - allowing imaging facilities to enjoy state-of-the-art imaging capabilities on their existing systems while saving capital expenditure in today's cost-constrained environments, including HDxt, MR Classic models such as 750, and at the same time, for the traditional 60cm magnetic resonance sold previously, an Evo upgrade solution has been launched. Without the need to replace magnets, it can be upgraded to its latest large-aperture Premier Evo system in situ.
Last May, Waukesha, Wisconsin-based GE Healthcare announced that AIR Recon DL is a deep learning image reconstruction technology that can be used in all anatomical structures and is now licensed by the U.S. FDA 510(k) on the Signa 7.0T MRI scanner.
GE Healthcare's deep learning image reconstruction technology - AIR Recon DL has been approved for its ultra-high field SIGNA™ 7T magnetic resonance scanner
Garry Gold, MD, professor of radiology at Stanford University and clinical reviewer of GE Healthcare's FDA application. "The images of are really amazing." AIR Recon DL for Signa 7.0T takes full advantage of our favorite ultra-high field 7.0T to visualize high-resolution tissue structures and take it to the next level by reducing noise and edge ringing. Therefore, AIR Recon DL with 7.0T helps uncover more comprehensive situations, provide better clinical insights for improving patient outcomes, and open up new opportunities for research in a variety of nursing fields. ”
