- This event has passed.
Distilling Crude EHR Data
November 10 @ 9:00 am - 10:30 am
Creating a clinical data repository for research is a challenging yet extremely useful task. By preemptively standardizing, normalizing, and transforming EHR data into a clinical data repository, thousands of hours can be saved when compared to working with raw EHR data. Via utilization of the Unified Medical Language System (UMLS) ontology, coding-agnostic queries can be written to extract more meaningful data much quicker and reliably than ever before. Because Cleveland Clinic uses this framework, live population exploration can occur right in front of clients where both population size and quality issues can be identified immediately.
Alex Milinovich is a System Analyst within the Quantitative Health Sciences (QHS) department at Cleveland Clinic. He is often referred to as the “data ninja” due to his extensive knowledge of data and how to find, map, clean, steal, query, interpret and generate it. He is involved directly with feasibility studies, project planning, cost estimation, data mining, data analysis and programming in a variety of programming languages. He regularly consult on data extraction from the electronic medical record and other sources for a large majority of the QHS outcomes projects that require data. This ranges from locating data to parsing and natural language processing of free-text notes to cleaning, mapping and validating the data. He has contributed to over 350 data analysis projects which yielded millions of dollars of both internal and external funding for the department as well as thousands of hours saved by pre-staging EHR data.