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

Novel Technique May Significantly Reduce Breast Biopsies
News | Breast Biopsy Systems | January 17, 2019
A novel technique that uses mammography to determine the biological tissue composition of a tumor could help reduce...
NewYork-Presbyterian Hospital Partners With Philips for Health IT and Clinical Informatics
News | Enterprise Imaging | January 16, 2019
Philips announced that NewYork-Presbyterian Hospital has chosen to implement the company’s IntelliSpace Enterprise...
Digital Mammography Increases Breast Cancer Detection
News | Mammography | January 16, 2019
The shift from film to digital mammography increased the detection of breast cancer by 14 percent overall in the United...
Artificial Intelligence Used in Clinical Practice to Measure Breast Density
News | Artificial Intelligence | January 15, 2019
An artificial intelligence (AI) algorithm measures breast density at the level of an experienced mammographer,...
Sponsored Content | Videos | Artificial Intelligence | January 15, 2019
ITN Contributing Editor Greg Freiherr offers an overview of...
Machine Learning Uncovers New Insights Into Human Brain Through fMRI
News | Neuro Imaging | January 11, 2019
An interdisciplinary research team led by scientists from the National University of Singapore (NUS) has successfully...
Videos | Interventional Radiology | January 11, 2019
Julius Chapiro, M.D., research faculty member and an...
AI Approach Outperformed Human Experts in Identifying Cervical Precancer
News | Digital Pathology | January 10, 2019
January 10, 2019 — A research team led by investigators from the National Institutes of Health and Global Good has de
Mobile App Data Collection Shows Promise for Population Health Surveys
News | Population Health | January 10, 2019
Mobile app data collection can bring access to more potential clinical study participants, reduce clinical study...
Artificial intelligence, also called deep learning and machine learning, was the hottest topic at the 2018 Radiological Society of North America (RSNA)) meeting.

Artificial intelligence was the hottest topic at the 2018 Radiological Society of North America (RSNA)) meeting, which included a large area with its own presentation therater set asside for AI vendors.

Feature | Artificial Intelligence | January 10, 2019 | Dave Fornell, Editor
Hands down, the hottest topic in radiology the past two years has been the implementation of...