Accelerated visualization, depicting 3D cardiac anatomy, function, and flow analysis, with cloud-based image processing, are key advances of ViosWorks, a major innovation for cardiac MRI. Although ViosWorks is not yet commercially available, GE Healthcare and the San Francisco start-up Arterys showcased it at the last ECR and RSNA meetings. It is expected to launch next fall.
“ViosWorks is extremely simple,” said GE Healthcare’s Ioannis Panagiotelis, Director of Global Marketing MRI, in an interview with MedicalExpo. The cardiac MRI can be reduced from more than 60 minutes to 6-to-10 minutes and the technologist’s work is reduced to just locating the chest. “Patients do not need to do breathholds as is done with conventional MRI and the 3D acquisition eliminates operator dependence,” Panagiotelis added. The advanced system also removes time-consuming slice positioning that is more error prone from the MR exam. Altogether, “it will enable general use of cardiac MRI economically.”
The cardiac MRI can be reduced from more than 60 minutes to 6-to-10 minutes and the technologist work is reduced to just locating the chest.
GE Healthcare has had a long-term relationship with Stanford University. Shreyas Vasanawala, MD, Chief of Body MRI and Co-Director of MRI at Stanford, and a Founder of Arterys, has been involved in developing and testing the platform.
“What this offers is a way of encoding all of the diagnostic information at once, permitting much quicker exams, no longer requiring special skills, and the quality of the information is much more reliable” than with conventional studies, he told MedicalExpo. ViosWorks shows every time point in the cardiac cycle, as well as time-resolved images of the beating heart, and a measure of the speed and direction of blood flow at each location.
3D cardiac anatomy, function, and flow in one free-breathing, 8 min scan (Courtesy of GE Healthcare)
Seven Dimensions of Data
The cloud-based imaging analytics platform, developed by Arterys, is designed to acquire seven dimensions of data, which include 3D heart anatomy, blood flow rate, and three in velocity direction of blood flow. From collecting thousands of images from physicians and radiologists, the system uses deep learning, a subset of artificial intelligence, to develop new algorithms. In moving from 2D to imaging 3D, dynamics and function, as much as 20 gigabytes of data are required.
The ViosWorks system uses deep learning, a subset of artificial intelligence, to develop new algorithms.
“You would need a supercomputer to process this and it’s why we are using the cloud,” said Panagiotelis. Data processing time is dramatically reduced. The platform uses new algorithms on large datasets never before incorporated that are reviewed in real time through cloud-based software developed by Arterys.
Arterys’ Chief Technology Officer John Axerio explained: “In partnering with GE and research physicians, we were able to build an incredibly rich database about the patient’s heart that doctors can use retrospectively and in planning treatment.”
“We’ve gotten a great reception from doctors and technologists because with ViosWorks, with one click, we start an MRI scan of the entire chest. Radiologists do not need to be in the room,” he said. The rapid image acquisition and the machine learning are features that add particular value, according to Axerio. “We are getting outstanding images in 3D, when 2D was the standard before.”