Turn on cookies to sign in
Signing in saves your progress to your Curie account. We can only do that with cookies on — turn them on to continue.
A new machine learning model can flag acute coronary syndrome patients at high risk for cardiogenic shock nearly two hours before hypotension sets in. Developed using admission data from 12,000 patients, the model achieved an AUROC of 0.84 in external validation. Researchers say this early warning window could prompt life-saving interventions before a patient's condition spirals.
A machine learning model developed at Northwestern Medicine can identify acute coronary syndrome (ACS) patients at high risk for cardiogenic shock a median of 1.8 hours before hypotension begins — potentially giving clinicians a critical window to intervene. Presented at the CHEST Annual Meeting, the model draws on admission data including demographics, vital signs, and lab values to generate a risk estimate the moment a patient arrives.
The tree-based model was trained on 8,400 ACS patients and externally validated on 3,600 more. Notably, it held up well outside its home institution — a common stumbling block for prediction models — retaining strong discriminatory performance in external testing.
Researchers are clear that this is a decision-support tool, not a replacement for clinical judgment. A high-risk alert is meant to prompt closer monitoring and earlier triage conversations, not automatic treatment.
By the Numbers:
Why it matters: Cardiogenic shock complicates 5–10% of ACS cases and carries a mortality rate exceeding 40%. A validated, real-time early warning tool embedded in the EHR could shift care from reactive to proactive — potentially saving lives in one of cardiology's most time-sensitive emergencies.
