Looking at an ultrasound image, YiRang Shin points to a region of the brain showing strong microbubble activity from a nerve stimulation test. [Image courtesy of Elizabeth Bello, Beckman Communications Office]
The team at the Beckman Institute for Advanced Science and Technology said the new technique enhances ULM, a diagnostic tool for microvascular imaging, utilizes deep learning. They call it localization with context awareness ultrasound localization microscopy, or LOCA-ULM. The team published their research in the journal Nature Communications.
ULM works by injecting FDA-approved microbubbles into blood vessels, where they act as contrast agents. Ultrasound waves then pinpoint the location of the microbubbles as they travel through the bloodstream by penetrating deep tissues. Researchers can use microbubbles to track blood flow speed and create spatial images of blood vessels.
The researchers said ULM’s imaging speed limited its use as a diagnostic tool in medicine and as a research tool in science. The new method created by the Beckman team delivers higher imaging performance and processing speed, plus increased sensitivity. According to the team, it also demonstrates improved computational and microbubble localization performance and is adaptable to different microbubble concentrations.
The researchers created a simulation model based on a generative adversarial network (GAN). It creates realistic microbubble signals to train the neural network, making localization faster, more accurate and more efficient.
“I’m really excited about making ULM faster and better so that more people will be able to use this technology. I think deep learning-based computational imaging tools will continue to play a major role in pushing the spatial and temporal resolution limits of ULM,” said first author YiRang Shin, a graduate student in the Department of Electrical and Computer Engineering at the University of Illinois Urbana-Champaign.
