Prognos manages the largest collection of clinical lab test records in the United States, with access to more than 25 billion records. From this data, they are able to get real-world insight to understand the journey and risk profile of patients, which they can translate into improved clinical practice and therapeutic development. To do this, Prognos has built a set of automated data pipelines and over 12,000 custom machine learning algorithms on the Databricks Unified Data Analytics Platform on AWS.
Attendees will learn how Prognos created:
- Streamlined data pipelines that enable different departments to utilize common components to achieve consistent results, while maintaining flexibility and control over their own workflows if needed
- Over 70 functions to refine, standardize and enrich clinical lab data configured with 162,905 rules
- Over 600 functions to clinically interpret medical lab data configured with 6,325 rules
- A unified data and analytics platform leveraging Apache Spark™, Databricks and AWS that:
- Facilitates collaboration between data scientists, data engineers, and analysts
- Enables teams to go directly from development to deployment
- Removes the overhead of dev ops tasks, allowing each team to focus on their speciality, increasing their productivity by 100%
This webinar will also include a live demo of an ML model to predict the cost of a patient’s visit with real-world data.
Adam Petranovich is the Chief Data Scientist at Prognos, managing a team of data scientists focused developing cutting edge algorithms that inform Prognos’ Clinical Truths™. Prior to joining Prognos, he spent 5 years in AdTech working as a Data Scientist and Engineer at AppNexus developing and implementing algorithms at scale. He also founded and served as CTO for NLP/AI startup Centairo, a real-time document discovery platform. Adam has broad experience developing a variety of predictive algorithms, feedback control systems, recommendation engines and language processing systems. Adam holds degrees in finance and economics from the University of Oregon.
Frank is the Technical Director for the Healthcare and Life Sciences vertical at Databricks. Prior to joining Databricks, Frank was a lead developer on the Big Data Genomics/ADAM and Toil projects at UC Berkeley, and worked at Broadcom Corporation on design automation techniques for industrial scale wireless communication chips. Frank holds a PhD and Masters of Science in Computer Science from UC Berkeley, and a Bachelor’s of Science with Honors in Electrical Engineering from Stanford University.