AI model HIPNO infers patient hemodynamics from non-invasive signals

Researchers have developed HIPNO (Hemodynamic Inference via Physics-informed Neural Operators), a novel AI model designed to non-invasively infer hemodynamic states from ubiquitous signals. HIPNO addresses a scale symmetry problem in physics-informed inference by parameterizing the network in its quotient space, using coordinates like compliance-normalized flow and decay time constant. Tested on over 945,000 intraoperative windows from 2,562 patients, HIPNO demonstrated a 32% lower error in predicting vascular decay compared to a population baseline, while maintaining accuracy in mean arterial pressure. AI

IMPACT This research could lead to more accessible and advanced hemodynamic monitoring in clinical settings, improving patient care.

RANK_REASON The cluster describes a new AI model presented in an arXiv paper for a specific scientific application. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• arXiv
• Hemodynamic Inference via Physics-informed Neural Operators
• patient
• windkessel effect

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →