Healthcare leaders across the U.S. are examining the most practical and beneficial ways to use artificial intelligence for clinical care beyond the typical narrow approach to solve specific problems.
During a panel at the World Medical Innovation Forum on May 11, healthcare executives outlined the most useful areas of artificial intelligence in the future. The panel which included Keith Dreyer, DO, PhD, clinical data science officer at Boston-based Partners and vice chairman of radiology at Massachusetts General Hospital in Boston, Mike Devoy, executive vice president of medical affairs and CMO of Bayer, and Karen DeSalvo, MD, chief health officer of Google Health, answered the question: What is on your AI wish list?
Dr. Keith Dreyer: If you think about what AI does today, it's typically narrow AI to be successful. That means there is a tremendous amount of data that is very consistently structured to solve a specific problem. If you take COVID as an example, a new disease that we are discovering the symptoms for, and discovering the treatments for and watching hundreds of thousands of people being afflicted with this, the key is to capture as much information as you can in a consistent way and start to look for things like how to potentially prevent the disease by social distancing; how to determine whether someone has the disease by looking at symptoms that are evolving; and also how to be able to monitor that patient through the care process and even prognose what it is that will happen to the patient specifically with the aggregated data that you had before. So I would say monitoring all the way through and as you watch these peaks going from start to finish and then start to come up again, that is when we start to address other areas like contact tracing like we are today.
Dr. Mike Devoy: I ascribe to the areas that Keith referred to and will add one. When we have this mobile disease? How do we look at treatments we have for other diseases? How can we repurpose drugs in our armamentarium because they will come much sooner than developing drugs from scratch?
Dr. Karen DeSalvo: I'll add two things: one, consumer facing, which has been our priority at Google, which is to make sure people have the right information at the right time that they need about COVID so they can take the actions to respond to social distancing or wash their hands or take care of their mental health. These are ways that the consumer and the community engage with the scientific and public health community. We believe that saves a lot of lives and has been a priority. I would put on the other end, supports for vaccine development and vaccine deployment. I think there will be a huge lift for the world to see that we can do this in a very short term way for speed, as the FDA says, but also do it in a way that drives equity across the world. This will take the best tools we have, and AI can help us with the best deployment for vaccines. Consumers and vaccines add nicely to their list of counter measures and what we need to do for public health and the healthcare system.