Live on Bloomberg TV

00:00BUT IT WAS ONLY RANKED EIGHT THREE OUT OF FIVE IN SERIOUSNESS. NOW I WANT TO BRING IN THE CHIEF INNOVATION OFFICER AT BOSTON CHILDREN ' S HOSPITAL. HOW DOES THIS AI MAPPING SYSTEM WORK QUESTION MARK? > > WHAT IS CHALLENGING ABOUT PUBLIC HEALTH ISSUES IS GETTING ENOUGH DETAIL IN REAL ABOUT WHAT IS HAPPENING ON THE GROUND. THERE IS ONLY SO MUCH PUBLIC HEALTH AGENCIES CAN DO AND FOR HEALTH SYSTEMS TO BE ABLE TO TRACK THESE CASES. IF YOU THINK ABOUT THE DIGITAL EXHAUST, OUR INTERACTIONS WITH SOCIAL MEDIA, SEARCHES ONLINE, ALL OF THAT INFORMATION IS INCREDIBLY VALUABLE BUT IT IS A HUGE MESSY SET OF DATA. IF WE START TAPPING INTO THAT DIGITAL EXHAUST AND GET INSIGHTS THROUGH MACHINE LEARNING AND AI, YOU CAN PINPOINT EARLY EVENTS BEFORE THEY BUBBLE UP TO OFFICIALS. WITH CORONAVIRUS, THE EARLIEST INSIGHTS CAME OUT OF SOCIAL MEDIA. THEY WERE TALKING ABOUT THE OUTBREAK THAT WAS ONLY A DOZEN PEOPLE OUT OF EIGHT SEASON MARKET THAT ULTIMATELY BECAME THE MAJOR EVENT WE ARE SEEING TODAY. WHAT WE ARE ABLE TO SEE IS THE SMALL SIGNALS, THEY WOULD BE CHALLENGING TO LOOK AT FROM ANY GOVERNMENT ENTITIES. > > WHAT IS THAT TELL YOU ABOUT THE CORONAVIRUS NOW? > > WE HAVE BEEN USING OUR TOOL NOT JUST TO IDENTIFY THE OUTBREAK BUT TO TRACK IT AS IT SPREAD. IF WE CONSTRUCT A MAD COW CASES AS THEY ARE IDENTIFIED ACROSS THE GLOBE, WE CAN QUICKLY SEE THE INFORMATION MODELS, MAKE PROJECTIONS AND UNDERSTAND FATALITIES ARE IT WE ARE SEEING A HUGE AMOUNT OF CASES THAT ARE EMERGING THAT MAY BE HAVE NOT INTERACTED WITH HEALTH SYSTEMS. WE KNOW THERE ARE A LOT OF CASES THAT ARE ASYMPTOMATIC. WE ARE TRYING TO UNDERSTAND THE SPREAD IN REAL-TIME TO PREPARE GOVERNMENTS FOR THE INEVITABLE ARRIVAL OF THE VIRUS. PAUL: IN TERMS OF USEFULNESS, EMILY POINTED OUT THAT THE VIRUS WAS ONLY RATED 83 OUT OF FIVE AND SERIOUSNESS. ARE THERE STILL SOME IMITATIONS TO THIS? > > ABSOLUTELY. AI IS JUST A SERIES OF TOOLS THAT THE PUBLIC HEALTH HAS. IT IS REALLY ABOUT THE FRONTLINE WORKERS. THEY NEED TO PROCESS THE INFORMATION. WE WERE COMMUNICATING EARLY ON WITH THE DEBBIE AJO ABOUT THESE EVENTS CREATED THEY ULTIMATELY MAKE THE DECISIONS ABOUT WHAT KIND OF RESPONSE TO TAKE ON. YOU HAVE TO UNDERSTAND THERE ARE MANY OUTBREAKS THAT HAPPEN ON A DAILY BASIS THAT ARE IDENTIFIED. THE CHALLENGE OF LOOKING AT THIS EVENT AND DECIDING IF THEY COULD TURN INTO A GLOBAL PANDEMIC, IT IS A REAL CHALLENGE BUT IT SHOWS THAT THESE TOOLS ARE IDENTIFYING SIGNALS GETTING COMMUNICATED AND WE ARE SEEING THIS IN CHINA AND THE WORLD STAGE. PAUL: TO YOUR POINT, BEFORE WE STARTED TODAY, I WAS PLAYING WITH HEALTH NOT -- HEALTH NET AND I SAW THAT THERE WAS AN OUTBREAK OF AVIAN FLU. > > WITHIN THE U.S., WE ARE DEALING WITH INFLUENZA. WE KNOW THAT IT LEADS TO CREDIBLE MORTALITY. THERE IS STILL A PUSH TO GET PEOPLE TO VACCINATE. THERE IS A MAJOR ABILITY FOR US TO HANDLE A NEW VIRUS PLUS THE SEASONAL VIRUSES THAT HAVE SOMETIMES A GREATER IMPACT. THE POINT OF THESE TOOLS IS TO PROVIDE TRANSPARENCY, HOPEFULLY ENCOURAGE PEOPLE TO TAKE THE RIGHT PRECAUTIONS REGARDLESS OF IF IT ' S CORONAVIRUS CREATED THE IDEA OF HAND WASHING, PROPER HYGIENE, ALL OF THESE THINGS ARE BASED ON PUBLIC HEALTH. YOU SHOULD BE PRACTICING REGARDLESS OF WHAT KIND OF VIRUS IS AROUND THE CORNER. > > AT THE SAME TIME THIS TECHNOLOGY HOLDS PROMISE, WHAT ABOUT THE LIMITATIONS OF ARTIFICIAL INTELLIGENCE? AND A RAPIDLY EVOLVING DISEASE ON THE GROUND. > > AI CAN ONLY DO SO MUCH TO UNCOVER SIGNALS, UNDERSTAND THE RELATIVE SEVERITY. WHAT WE REALLY NEED US HUMANS IN THE LOOP. WE ARE WORKING NOT JUST WITH MACHINE LEARNING TECHNOLOGY BUT WITH NATIONAL AND INTERNATIONAL EXPERTS. THEY ARE LOOKING AT THIS INFORMATION, PROCESSING AND CONTEXTUALIZING AND HELPING TO MAKE THEIR DECISIONS. WE CAN UNCOVER MASSIVE AMOUNTS OF DATA THROUGH THESE SOCIAL NETWORKS AND PROCESS IT AND PUT IT IN THE HANDS OF EXPERTS BUT IT IS ULTIMATELY DOES EXPERTS TAKING THE SITUATIONAL AWARENESS AND MAKING THIS -- THESE CRITICAL DECISIONS. IT IS NOT ABOUT AI REPLACING A PUBLIC HEALTH AGENCY BUT AUGMENTING WITH THE BEST POSSIBLE DATA IN REAL-TIME. > > WHEN YOU LOOK AT THE DATA, DO YOU SEE POTENTIAL HOTSPOTS THAT WE MIGHT NOT BE AWARE OF? THAT GOVERNMENTS NEED TO KNOW ABOUT. > > ULTIMATELY, WHAT WE ARE SEEING IS BUBBLING UP CASES ACROSS THE MIDDLE EAST AND NOW STARTING IN EUROPE. IT IS NOT TOO DISSIMILAR FROM WHAT WE ARE SEEING THROUGH JUDICIAL SOURCES. WHAT WE ARE INTERESTED IN IS PICKING UP THE EARLIER INSIGHTS OF WHAT MIGHT HAPPEN IN THE U.S.. THAT ' S WHY WE ARE TAPPING INTO NEW DATA STREAMS READ CROWDSOURCING, SEARCH QUERIES, THE IDEA THAT WE ARE NOW LOOKING AT FINE-GRAINED DETAIL DATA TO SEE IF WE ARE GOING TO MISS THE CLUSTERS OF CASES THAT WILL POP UP WITHIN THE U.S.. THAT IS THE FOCUS RIGHT NOW. ARE WE GOING TO BE ABLE TO SEE THE INCREASE IN CASES THAT WILL LIKELY HIT THE U.S. AT THE EARLIEST OUTSET SO WE CAN PREPARE THE POPULATION? WE KNOW THE PUBLIC HEALTH AND HEALTH SYSTEMS ARE GOING TO BE INCREDIBLY OVERWHELMED AND POSSIBLY CASES WILL BE MISSED. WE KNOW THE TRAVEL HISTORY CONCEPT IS REALLY NOT A SOLID BECAUSE WE HAVE SEEN CASES POP UP AROUND THE GLOBE. YOU CAN ' T THINK ABOUT TRAVEL HISTORY AS A WAY TO UNDERSTAND. WE ARE SURVEYING THE POPULATION THROUGH A TOOL WE HAVE DEVELOPED AND OTHERS THAT HOPEFULLY WILL GIVE US THE EARLY INSIGHT FOR WHEN THE INEVITABLE EVENT HAPPENS. PAUL: THANK YOU SO MUCH FOR JOINING US. STILL TO COME, GRUBHUB ROLLS OUT

No schedule data available.

Boston Children's Hospital Uses Data to Create an AI Coronavirus Map

John Brownstein, Boston Children's Hospital chief innovation officer, discusses how the hospital made an AI tool to map the spread of the coronavirus. He speaks on "Bloomberg Markets." (Source: Bloomberg)