Hemodynamics, fluids & transfusion
Deep learning model to identify and validate hypotension endotypes in surgical and critically ill patients
Authors' conclusion
Unsupervised deep learning identified four endotypes of hypotension in surgical and critically ill patients: vasodilation, hypovolaemia, myocardial depression, and bradycardia.
Study context
- Population
- Surgical and critically ill patients with hypotension (MAP <65 mm Hg for at least 1 min)
- Outcome
- Endotypes of hypotension based on stroke volume index, heart rate, systemic vascular resistance index and stroke volume variation during hypotensive episodes
Methodological detailsdeep learning autoencoder combined with Gaussian mixture model, two independent validation datasets
- Deep learning autoencoder combined with Gaussian mixture model
- Two independent validation datasets
Performance
As reported in the abstract
Hypotension endotypes (development dataset: 871 surgical patients, 6962 hypotensive events)
- Four hypotension endotypes identified: vasodilation, hypovolaemia, myocardial depression, bradycardia
- 4
Further findings
Hypotension endotypes in validation datasets (1000 surgical patients, 7904 hypotensive events; 1000 critically ill patients, 53 821 hypotensive events)
- Same four hypotension endotypes identified in both independent validation datasets
- 4
Original publication
Deep learning model to identify and validate hypotension endotypes in surgical and critically ill patients. Br J Anaesth. Published 8 January 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →