Using AI and Collaborative Workflows to Predict and Prevent Clinical Deterioration

Stanford Medicine embarked on a mission to bridge the gap in utilizing machine-learning (ML) models for enhancing care delivery. They focused on three fundamental challenges within the health system:

1. Establishing a comprehensive framework for integrating artificial intelligence (AI) into complex healthcare workflows.
2. Assembling and developing teams comprising individuals, technologies and processes essential for the successful creation and deployment of AI-driven systems.
3. Implementing these changes in a sustainable and scalable manner suitable for the healthcare enterprise.

This case study serves as a real-world example of the initial stages of implementing AI in care delivery. Specifically, the focus was on predicting clinical deterioration to minimize unplanned transfers to the intensive care unit (ICU). We’ll delve into the application of design principles to the health system, discuss the obstacles faced, as well as the factors that facilitated progress. These experiences show a collaborative strategy for harnessing AI to enhance patient outcomes and safety.