Hemodynamics, fluids & transfusion
Blood Pressure Estimation Using Explainable Deep-Learning Models Based on Photoplethysmography
Authors' conclusion
Our deep-learning model noninvasively estimates arterial pressure with high accuracy.
Study context
- Population
- Patients in neuroradiology and digestive surgery with arterial linemedian 2201 measurements per patient
- Participants
- 117 patients
- Comparison
- Deep-learning model vs mean arterial pressure (MAP) baseline model
- Outcome
- Mean absolute error (MAE) of MAP vs arterial line
Methodological detailsdeep-learning model development, external test set
- Deep-learning model development
- External test set
Performance
As reported in the abstract
MAE of MAP
- Improvement of deep-learning model over baseline
- 42.6%
Further findings
Explainability (Grad-CAM saliency map)
- Model primarily extracts information near the dicrotic notch region
Original publication
Blood Pressure Estimation Using Explainable Deep-Learning Models Based on Photoplethysmography. Anesth Analg. Published 16 December 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →