Key result
Patient-specific neural network using four PPG features enables automated beat-to-beat blood pressure estimation.
Why the study?
Most reported generalized PPG-based BP estimation methods often lack desired accuracy due to pathophysiological diversity and rely on non-globalized correction factors.
Does a patient-specific neural network model using PPG features accurately estimate blood pressure in ICU patients?
Population
670 records of 50 ICU patients from MIMIC, MIMIC II and MIMIC Challenge databases
Comparison
Patient-specific neural network modeling vs generalized BP estimation models
Authors
Loading...
May support cuffless BP monitoring in select ICU patients; leaves open prospective validation before practice change.
Does a patient-specific neural network model using PPG features accurately estimate blood pressure in ICU patients?
A patient-specific neural network approach using PPG signals enables accurate, cuff-less beat-to-beat blood pressure estimation without requiring global correction factors.
Chakraborty et al. (2020) studied Intensive care unit (ICU) patients (n=50). Patient-specific neural network (NN) modeling using PPG features vs. Generalized BP estimation models was evaluated on Accuracy of estimated systolic (SBP) and diastolic (DBP) blood pressure. A patient-specific neural network model using four photoplethysmography features enabled automated beat-to-beat estimation of systolic and diastolic blood pressure across 670 records from 50 patients.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: