This talk is neither about big data, nor about AI. It's about artisanal, handcrafted data that poses a real challenge for anyone trying to analyze it: healthcare data. In theory, applying AI to healthcare sounds like the perfect match - look at real world data generated by patients, apply AI, learn from trends, and improve healthcare outcomes based on those learnings.
Systems like IBM Watson make us believe that the problem is already solved, but in reality, real world healthcare data and its applications suffer from problems not encountered in other domains, which poses huge challenges for any kind of analytical applications. In this talk, we will look at the landscape of messy and patchy healthcare data, understand the difficulties of drawing reasonable conclusions from the data, and discuss the challenges of changing user behavior in healthcare.
Sam Bail is a Data Insights Engineer/Data Scientist in New York with a passion for data for social good. After spending several years in Academia researching ontologies, OWL, and the Semantic Web, Sam became an early employee at Flatiron Health, a New York City based healthcare technology startup, where she helped build up many of the company's data products. Sam is the co-founder of Manchester Girl Geeks, a UK-based not-for-profit organization for girls and women in STEM, as well as a contributor to NYC PyLadies and Women Who Code.