Blog

From machine learning to decisions

Nov 7, 2017

The goal of data science in health care is to build systems which make use of data to help patients and physicians make the best possible decisions about their health. One way that we have approached this challenge at Penn Medicine Predictive Healthcare is to employ machine learning to predict which patients are at elevated risk of adverse events (for example sepsis, readmissions) at any given moment. The resulting algorithms, built on a large collection...

Read More >

Some challenges in creating predictive healthcare solutions

Jul 20, 2017

Here at Penn Medicine, we’re developing uses of AI and machine learning to help improve medical decision making. As the electronic medical record (EMR) has matured, and its adoption becomes widespread, there has been an increasing amount of data available to work with toward this end. However, this data comes with several unique challenges – many of which we have encountered firsthand, and others that we expect to face soon. Some aspects of these are...

Read More >