The science behind self-driving cars seems to have found a purpose in radiology. A group of interventional radiologists at University of California, Los Angeles are using a chat application to streamline and speed-up diagnosis and treatment decisions. This innovative use of artificial intelligence ultimately improves patient-doctor communication by utilizing a variety of programs, infographics, and pre-designed responses to relay information to between doctors and patients through a chat interface.
Chat programs, which were the life-force of the internet during the early 2000s, have made a comeback in the science of deep learning. In the case of the UCLA study, the researchers have programmed an application with more than 2,000 example data points that resemble common consultation inquiries, akin to online customer service chats or AOL Instant Messenger. The app has “learned” to produce answers to questions in a “chat” like format. The more physicians use the app, the more information it acquires, gradually enhancing its intelligence.
"This research will benefit many groups within the hospital setting,” wrote Kevin Seals, MD, one of the app’s programmers, resident physician in radiology at UCLA, and the study’s co-author. “Patient care team members get faster, more convenient access to evidence-based information; interventional radiologists spend less time on the phone and more time caring for their patients; and, most importantly, patients have better-informed providers able to deliver higher-quality care.”