Interventional radiologists at UCLA have created a virtual radiology consultant to provide clinical decision support for other medical providers. Using natural language processing with IBM’s Watson, the team developed the chatbot by giving it more than 2,000 data points that simulated common queries interventional radiologists receive during a consultation. The result is a virtual assistant that can answer a clinician’s question via text in a matter of seconds to help them select the best course of action. “It’s like texting with a human radiologist, but it uses artificial intelligence to automatically respond. It’s the most rapid way to get information, and it is very carefully curated information supported by medical data, so the result is better patient care,” said Kevin Seals, M.D., resident physician in radiology at UCLA and the application programmer.
A Faster Response and Better Patient Care
The virtual assistant can’t answer every question, but it can answer routine ones. For instance, it can tell a provider whether to perform a specific contrast on a patient with a specific allergy. And if the chatbot believes the query needs a human response, it can provide the contact information of an interventional radiologist in the hospital. This speedy response for common questions will save time and allow interventional radiologists to spend more time on patients who require it. “We hope it might free up time for radiologists to focus on patients and take care of the more complex situations that can’t be automated,” Seals says. While the program is still being tested, the team at UCLA hopes it will go live in the next month.