by
Lauren Dubinsky, Senior Reporter | March 13, 2017
The same technology employed in self-driving cars has the potential to help interventional radiologists. Researchers at the University of California at Los Angeles leveraged deep learning to create a tool that can communicate with referring clinicians and rapidly answer common questions.
"I think it will eventually be the most seamless way to share medical information,” Dr. John Hegde, resident physician in radiation oncology at UCLA, said in a statement. “Although it feels as easy as chatting with a friend via text message, it is a really powerful tool for quickly obtaining the data you need to make better-informed decisions."
If this tool is eventually incorporated into clinical practice, interventional radiologists will spend less time on the phone and more time with their patients, and referring clinicians will gain faster, more convenient access to evidence-based information.

Ad Statistics
Times Displayed: 110001
Times Visited: 6644 MIT labs, experts in Multi-Vendor component level repair of: MRI Coils, RF amplifiers, Gradient Amplifiers Contrast Media Injectors. System repairs, sub-assembly repairs, component level repairs, refurbish/calibrate. info@mitlabsusa.com/+1 (305) 470-8013
Deep learning refers to networks of artificial neurons that evaluate large data sets to automatically find patterns and “learn” without human intervention. These artificial networks can offer information pertaining to early detection, treatment planning, and disease monitoring.
The tool has been fed over 2,000 data points that simulate inquiries interventional radiologists typically receive during a consultation. It instantly provides the best answer to the referring clinicians, so they can inform patients in real-time about the next phase in treatment.
The answers are presented in a format similar to online customer service chats and the information can be in the form of websites, infographics or custom programs.
If the tool determines that a human is needed to provide a response, the clinician is given contact information for an interventional radiologist. As clinicians use the tool, it “learns” from each scenario and gradually becomes more intelligent.
The tool can answer questions that are posed in natural language because it’s based on a technology called Natural Language Processing, which was implemented using IBM’s Watson.
Research that investigated this new tool was presented this week at the Society of Interventional Radiology’s 2017 annual scientific meeting. A small team of hospitalists, radiation oncologists and interventional radiologists at UCLA is now testing a prototype.
As the team continues to make improvements to the tool, it may eventually be capable of assisting general physicians in interfacing with other specialists, such as cardiologists and neurosurgeons.
Back to HCB News