News | Artificial Intelligence | July 17, 2018

Artificial Intelligence Provides Faster, Clearer MRI Scans

Machine learning software produces better signal and less noise than conventional MRI techniques

Artificial Intelligence Provides Faster, Clearer MRI Scans

A new artificial-intelligence-based approach to image reconstruction, called AUTOMAP, yields higher quality images from less data, reducing radiation doses for CT and PET and shortening scan times for MRI. Shown here are MR images reconstructed from the same data with conventional approaches, at left, and AUTOMAP, at right. Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital

July 17, 2018 — A research team with funding from the National Institute for Biomedical Imaging and Bioengineering (NIBIB) has developed an advanced computing technique for rapidly and cost effectively improving the quality of biomedical imaging. The technology, called AUTOMAP, uses machine learning and software, referred to as neural networks — inspired by the brain’s ability to process information and perceive or make choices. AUTOMAP finds the best computational strategies to produce clear, accurate images for various types of medical scans.

In their study in the March 21, 2018, issue of Nature, the researchers from Massachusetts General Hospital (MGH) Martinos Center for Biomedical Imaging and Harvard University found that the AUTOMAP system could produce brain magnetic resonance imaging (MRI) images with better signal and less noise than conventional MRI techniques. Achieving a good signal-to-noise ratio is a key factor in generating a quality MRI scan.

MRI uses a magnetic field and radio waves to create detailed images of tissues inside the body. Noise from either electronic sources or from tissues in the body are detrimental to image quality, so imagers look for ways to lessen its effect.

“The signal-to-noise ratio improvements we gain from this artificial intelligence-based method directly accelerates image acquisition on low-field MRI,” said lead author Bo Zhu, Ph.D., postdoctoral research fellow in radiology at Harvard Medical School and in physics at the MGH Martinos Center. NIBIB has supported Zhu’s postdoctoral research on this project. He added that the AUTOMAP neural network will be compatible with novel image acquisition strategies and unconventional hardware designs.

AUTOMAP churns through — and learns from — data from existing images and applies mathematical approaches in reconstructing new ones. The team used a set of 50,000 MRI brain scans from the NIH-supported Human Connectome Project to train the AUTOMAP system to reconstruct images in their study, successfully demonstrating improvements in reducing noise and reconstruction artifacts over existing methods.

AUTOMAP achieves almost instantaneous image reconstruction, according to senior author Matt Rosen, Ph.D., director of the Low-field MRI and Hyperpolarized Media Laboratory and co-director of the Center for Machine Learning at the MGH Martinos Center. The reason for the rapid processing speed — just tens of milliseconds — is that the neural network has no cycles or loops, rather is a feedforward system.

“Some types of scans currently require time-consuming computational processing to reconstruct the images,” Rosen said. “In those cases, immediate feedback is not available during initial imaging, and a repeat study may be required to better identify a suspected abnormality. AUTOMAP would provide instant image reconstruction to inform the decision-making process during scanning and could prevent the need for additional patient visits.”

“This technology could become a game changer, as mainstream approaches to improving the signal-to-noise ratio rely heavily on expensive MRI hardware or on prolonged scan times,” said Shumin Wang, Ph.D., director of the NIBIB program in magnetic resonance imaging. “It may also be advantageous for other significant MRI applications that have been plagued by low signal-to-noise ratio for decades, such as multi-nuclear spectroscopy.”

For more information: www.nature.com/nature

Reference

Zhu B., Liu J.Z., Cauley S.F., et al. "Image reconstruction by domain-transform manifold learning." Nature, March 21, 2018. https://doi.org/10.1038/nature25988

Related Content

Advanced technologies and applications such as point-of-care, pediatrics, dry-magnets, compact MRI and fusion imaging are driving global market
News | Magnetic Resonance Imaging (MRI) | February 24, 2021
February 24, 2021 — Frost & Sullivan's recent analysis, Technological Advancements and Emerging Applications in t
Detroit-based magnetic resonance imaging (MRI) technology company SpinTech, Inc. has acquired medical-imaging research and technology developer Magnetic Resonance Innovations, Inc. (MR Innovations).
News | Magnetic Resonance Imaging (MRI) | February 24, 2021
February 24, 2021 — Detroit-based magnetic resonance...
Findings indicate that PPC and GG are highly predictive of overall upstaging by PSMA PET/CT for patients with high-risk prostate cancer

Image courtesy of UCLA Health

News | PET-CT | February 23, 2021
February 23, 2021 — A...
icobrain cva allows the quantitative assessment of tissue perfusion by reporting the volume of core and perfusion lesion by quantifying Tmax abnormality and CBF abnormality together with the mismatch volume and ratio
News | Artificial Intelligence | February 23, 2021
February 23, 2021 — icometrix, world leader in imaging...
F-18 FES PET images of patients with ER+/PR+/HER2- invasive ductal carcinoma. Left panel: Progressive disease seen at the 8-week time-point in a patient on sequential therapy. Right panel: Stable disease through all 3 time-points, remaining on study therapy for 6.7 months until disease progression on combined vorinostat aromatase inhibitor therapy. Image created by Lanell M Peterson, Research Scientist, University of Washington Medical Oncology, Seattle WA.

F-18 FES PET images of patients with ER+/PR+/HER2- invasive ductal carcinoma. Left panel: Progressive disease seen at the 8-week time-point in a patient on sequential therapy. Right panel: Stable disease through all 3 time-points, remaining on study therapy for 6.7 months until disease progression on combined vorinostat aromatase inhibitor therapy. Image created by Lanell M Peterson, Research Scientist, University of Washington Medical Oncology, Seattle WA.

News | Molecular Imaging | February 22, 2021
February 22, 2021 — Molecular imaging
Axial FLAIR MR image shows T2 prolongation in bilateral middle cerebellar peduncles (arrows). Findings were associated with restricted diffusion and areas of T1 hypointense signal without enhancement or abnormal susceptibility. Image courtesy of American Roentgen Ray Society (ARRS), American Journal of Roentgenology (AJR)

Axial FLAIR MR image shows T2 prolongation in bilateral middle cerebellar peduncles (arrows). Findings were associated with restricted diffusion and areas of T1 hypointense signal without enhancement or abnormal susceptibility. Image courtesy of American Roentgen Ray Society (ARRS), American Journal of Roentgenology (AJR)

News | Coronavirus (COVID-19) | February 22, 2021
February 22, 2021 — According to an...
GE Healthcare introduced its artificial intelligence (AI) automation features on its Voluson Swift ultrasound platform at the 2020 Radiological Society of North America (RSNA) virtual meeting. Features of this system include semi-automated contouring, auto identification of fetal anatomy and positioning on imaging. AI is seeing increasing integration in ultrasound systems from numerous vendors.

GE Healthcare introduced its artificial intelligence (AI) automation features on its Voluson Swift ultrasound platform at the 2020 Radiological Society of North America (RSNA) virtual meeting. Features of this system include semi-automated contouring, auto identification of fetal anatomy and positioning on imaging. AI is seeing increasing integration in ultrasound systems from numerous vendors.

Feature | Ultrasound Imaging | February 18, 2021 | By Dave Fornell, Editor
Recent advances in ultrasound image sy...
Example MR images from paediatric brain tumour patients. This first column shows T1-weighted images following the injection of gadolinium contrast agent. The second column shows T2-weighted images and the final column shows apparent diffusion coefficient maps calculated from diffusion-weighted images. (a–c) are taken from a patient with a Pilocytic Astrocytoma, (d–f) are from a patient with an Ependymoma and (g–i) were acquired from a patient with a Medulloblastoma.

Example MR images from paediatric brain tumour patients. This first column shows T1-weighted images following the injection of gadolinium contrast agent. The second column shows T2-weighted images and the final column shows apparent diffusion coefficient maps calculated from diffusion-weighted images. (ac) are taken from a patient with a Pilocytic Astrocytoma, (df) are from a patient with an Ependymoma and (gi) were acquired from a patient with a Medulloblastoma. Image courtesy of Nature Research Journal

News | Pediatric Imaging | February 17, 2021
February 17, 2021 — Diffusio...