Epic Systems’s Sepsis Prediction Algorithms Revealed to Have High Error Rates on Seriously Ill Patients

August 1, 2021

University of Michigan Hospital investigators uncovered high rates of false positives and false negatives in Epic System's sepsis prediction algorithms. The findings challenge the company's published claims, emphasizing the need for trustworthy AI, responsible AI governance, and the importance of HISPI Project Cerebellum TAIM (Measure) to improve harm prevention.

For those interested in shaping the future of AI, JOIN US.

Matched TAIM controls

Suggested mapping from embedding similarity (not a formal assessment). Browse all TAIM controls

MEASURE 2.6 — similarity 0.706, rank 1. TAIM detail and related incidents →
MAP 3.2 — similarity 0.663, rank 2. TAIM detail and related incidents →
MEASURE 3.1 — similarity 0.663, rank 3. TAIM detail and related incidents →

Alleged deployer: university-of-michigan-hospital
Alleged developer: epic-systems
Alleged harmed parties: sepsis-patients

Source

Data from the AI Incident Database (AIID). Cite this incident: https://incidentdatabase.ai/cite/123

Data source

Incident data is from the AI Incident Database (AIID).

When citing the database as a whole, please use:

McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. In Proceedings of the Thirty-Third Annual Conference on Innovative Applications of Artificial Intelligence (IAAI-21). Virtual Conference.

Pre-print on arXiv · Database snapshots & citation guide

We use weekly snapshots of the AIID for stable reference. For the official suggested citation of a specific incident, use the “Cite this incident” link on each incident page.