H2O Driverless AI for CDS: Early Detection of Sepsis in the ICU

Presented by

Niki Athanasiadou MRes, PhD, Customer Data Scientist, H2O.ai

About this talk

Clinical decision support (CDS) systems are patient-focused alerts, reminders and clinical guidelines that help healthcare providers improve patient outcomes and enhance healthcare workflows. AI-backed CDS offers the opportunity for more ‘intelligent’ systems that can detect risk of disease more accurately and at an earlier time, when interventions might be more effective. In the use case presented in this webinar we will use H2O.ai’s award-winning automatic machine learning platform, H2O Driverless AI, to detect patient-specific risk of sepsis in Intensive Care Unit (ICU) six hours before it is actually diagnosed. As features, we will be using patient-specific vital signs, laboratory tests and basic demographic information, all typically available in the ICU. Finally, by delving into the advanced model explainablity capabilities that are available within Driverless AI, we will demonstrate how Driverless AI offers insights into possible paths for intervention that trained medical personnel can take advantage of. In this webinar, you will learn: - How to prepare ICU time-dependent data for machine learning. - How to handle imbalanced patient data to train accurate models for medical use. - How to leverage machine learning explainablity techniques for CDS.

Related topics:

More from this channel

Upcoming talks (0)
On-demand talks (111)
Subscribers (19201)
H2O.ai is the maker of H2O, the world's best machine learning platform and Driverless AI, which automates machine learning. H2O is used by over 200,000 data scientists and more than 18,000 organizations globally. H2O Driverless AI does auto feature engineering and can achieve 40x speed-ups on GPUs.