A Bridge Between Care and Code: An Inquiry into Human-Centred AI in Intensive Care Decision-Making

Abstract

This thesis investigates how Artificial Intelligence and Machine Learning (AIML)-based clinical decision support systems can be designed to align with intensive care clinicians’ values and needs. It takes a value-sensitive and tension-aware approach, showing how tensions in ICU decision-making signal complexity, potential trade-offs and limitations of AIML support. Insights gained from interviews and prototyping demonstrate that AIML based tools are most supportive when contextually anchored, integrated into existing workflows, and positioned as a tool that complements existing processes and devices. Within data science, this research offers a relational, context-sensitive and tension-aware approach of thinking for problem formulation.

Ähnliche Arbeiten