Speaker
Availability of the Speaker<br>Let us know if there are times you CANNOT present,<br>prehaps because you need to leave for the airport early, etc.
Available for any speaking time.
Summary (2-4 sentences)<br>Just a few informal sentences describing what you want to present.<br>No need to spend a lot of time on this! You can change it later.
The widespread digitization of patient data via electronic health records (EHRs) has created an
unprecedented opportunity to use machine learning algorithms to better predict disease risk at the
patient level. Although predictive models have previously been constructed for a few important
diseases, such as breast cancer and myocardial infarction, we currently know very little about
how accurately the vast majority of diseases can be predicted, and how far in advance.
Through the use of HTCondor and the CHTC cluster at UW-Madison, we build predictive models for nearly every ICD-9 code across several time points.
Here we show that such pan-diagnostic prediction is possible with a high level of performance across
diagnosis codes.
