Medical device data presents a major opportunity to improve patient care. By applying machine learning techniques to this data, hospitals can automate patient monitoring, even after specialists have left for the night. Furthermore, medical device companies can use this data to anticipate device errors, reducing costs through preventative maintenance. And finally, companies leveraging wearable device data can glean new insights on the human body outside of a clinical setting.

Despite the opportunity, medical device data introduces significant scale challenges. For example, a 1,000 bed hospital with a single medical device at each bed taking measurements at 1KHz will produce over 86 billion measurements per day. Ingesting and processing this data is a daunting task magnified by legacy data architectures. Attempting to aggregate these datasets across long time horizons and large patient cohorts to improve analyses is near impossible for most organizations. Marshaling data as it streams in from devices further complicates these problems.

In this webinar, we will walkthrough how to use Databricks Unified Analytics Platform and open source technologies to overcome these challenges and model medical device data at scale.

Join this technical session to learn how to:

• Build a streaming pipeline for EKG data using Structured Streaming and Delta Lake 
• Improve data consistency guarantees while eliminating data engineering bottlenecks
• Interactively query streaming EKG data in real-time 
• Rapidly train a deep learning model over terabytes of waveforms 
• Track and manage the entire model lifecycle in MLflow, allowing analysis traceability