Dr. Alex Jaimes
Tuesday, November 1, 2016 - 3:30pm
Doctors spend on average 14 years training before they are practice. Once they do, on average, they see about 2,000 patients a year. Each patient, in turn produces a trove of data, a very small percentage of which is captured during a doctor’s visit. That data, however, is multi-modal, consisting of unstructured text, structured text, etc. In many ways one could argue that a big portion of a doctor’s job encompasses making predictions based on data: during those years of training, doctors essentially build mental models that they use to later on make predictions (diagnoses). In spite of a long training period and of seeing so many patients, the numbers seen by any individual practitioner are rather small. In aggregate, however, healthcare produces huge amounts of data. In addition, vast amounts of non-healthcare data is being continuously collected, much of which is directly related to healthcare. Data on the movement of people and goods, for instance, can produce insights into the spread of disease. Environmental data, correlated with disease data, can be used to predict and prevent medical conditions within specific demographic and/or socioeconomic groups. In this talk I will describe how at Acesio we are integrating multiple types of data, at scale, to make healthcare more efficient and to obtain actionable insights at the practitioner, hospital, and macro scale. I’ll discuss particular challenges in healthcare, the phenomenal opportunities that exist, and how data will revolutionize medicine.