Artificial Intelligence

In a special report published Jan. 30 in the inaugural issue of Radiology: Artificial Intelligence, Luciano M. Prevedello, MD, and colleagues recognize the challenges of implementing AI into clinical workflows, but also offer potential solutions—specifically image-based competitions—which could foster faster, more collaborative AI research.

The Massachusetts Institute of Technology (MIT)’s Laboratory for Computational Physiology has published their MIMIC-Chest X-Ray Database (MIMIC-CXR)—a collection of more than 350,000 chest x-rays associated with 227,943 imaging studies sourced from Beth Israel Deaconess Medical Center in Boston.

The system, named Doctor Alzimov, after the Russian-born science fiction writer, can be installed on any computer and can find nodules as small as 2 millimeters.

In 2018, capital investments in startup companies developing medical imaging AI solutions reached almost $580 million—more than double the 2017 amount of $270 million, according to a new market report published Jan. 31 by Signify Research.

A 14-layer convolutional neural network (CNN) trained on MRI and pathology data accurately predicted the molecular subtype of breast cancers, according to a Jan. 31 study published in the Journal of Digital Imaging. The method may help personalize treatment plans for the disease.

The world market for digital pathology is expected to rise to $600 million by 2022, according to a new report from Signify Research. But it will have to overcome some strong obstacles first.

A team from Singapore demonstrated that an object detection convolutional neural network (CNN) could accurately detect and localize fractures on wrist x-rays, according to a Jan. 30 study published in Radiology: Artificial Intelligence. The method may be more verifiable than traditional CNNs.

RSNA has published the first issue of its new online journal, Radiology: Artificial Intelligence.

Researchers from Stanford University in California have published a large, public dataset containing more than 224,000 chest x-rays from more than 65,000 patients to train AI algorithms. The team also announced a competition inviting developers to submit their chest x-ray interpretation models to detect pathologies more accurately than certified radiologists. 

In an effort to destigmatize mental illness and help patients find better treatments, researchers from Virginia Tech’s Fralin Biomedical Research Institute in Roanoke, Virginia trained a machine learning algorithm with brain fMRI scans to diagnosis mental disorders more accurately than standard methods, according to a recent report by The Verge.

“Students rely on us to understand how radiology is incorporating new technology and what the future of the field will look like for them, but many of us are ill prepared to teach the younger generation about this, mostly because we ourselves are not sure,” Allison Grayev, MD, wrote in an editorial published in Academic Radiology. 

A combined deep learning method better detected hemorrhages and identified different subtypes of intracranial hemorrhage than single algorithms used alone, according to a new study published in the Journal of Digital Imaging.