Timezone: »

Medical Dead-ends and Learning to Identify High-Risk States and Treatments
Mehdi Fatemi · Taylor Killian · Jayakumar Subramanian · Marzyeh Ghassemi

Tue Dec 07 08:30 AM -- 10:00 AM (PST) @ None #None

Machine learning has successfully framed many sequential decision making problems as either supervised prediction, or optimal decision-making policy identification via reinforcement learning. In data-constrained offline settings, both approaches may fail as they assume fully optimal behavior or rely on exploring alternatives that may not exist. We introduce an inherently different approach that identifies "dead-ends" of a state space. We focus on patient condition in the intensive care unit, where a "medical dead-end" indicates that a patient will expire, regardless of all potential future treatment sequences. We postulate "treatment security" as avoiding treatments with probability proportional to their chance of leading to dead-ends, present a formal proof, and frame discovery as an RL problem. We then train three independent deep neural models for automated state construction, dead-end discovery and confirmation. Our empirical results discover that dead-ends exist in real clinical data among septic patients, and further reveal gaps between secure treatments and those administered.

Author Information

Mehdi Fatemi (Microsoft Research)
Taylor Killian (University of Toronto Vector Institute)
Jayakumar Subramanian (Adobe Systems)
Marzyeh Ghassemi (University of Toronto, Vector Institute)

More from the Same Authors