Patients are more likely to die when a human doctor overrides an AI's 'all-clear' signal to diagnose a blood clot.
April 1, 2026
Original Paper
Human-AI Collaboration in Radiology: The Case of Pulmonary Embolism
SSRN · 6482718
The Takeaway
We usually assume human experts should have the final word, but this study found that when radiologists overruled AI to diagnose a pulmonary embolism, patient mortality actually increased by 5.5%. This suggests that human 'catches' in these cases often represent overdiagnosis and lead to unnecessary, risky treatments.
From the abstract
We study how radiologists use AI to diagnose pulmonary embolism (PE), the third leading cause of cardiovascular death in the U.S. We track over 100,000 scans interpreted by nearly 400 radiologists during the staggered rollout of an FDA-approved diagnostic platform in a large hospital system. When AI flags PE, radiologists agree 84% of the time; when AI predicts no PE, they agree 97%. Disagreement evolves substantially: radiologists initially reject AI-positive PEs in 30% of cases, dropping to 12