Related Experiment Video
Updated: Aug 11, 2025

Software for Analysis of Heart Rate and Blood Pressure Time-series Data from the Valsalva Maneuver
Published on: June 27, 2025
Blood pressure estimation by spatial pulse-wave dynamics in a facial video.
Kaito Iuchi1,2, Ryogo Miyazaki1,2, George C Cardoso3
1Graduate School of Science and Engineering, Department of Imaging Science, Chiba University, Japan.
This study introduces a novel remote method to estimate continuous blood pressure (BP) using facial videos. The technique leverages spatial pulse-wave analysis via deep learning, achieving accurate BP estimation with significantly lower error than traditional methods.
Area of Science:
- Biomedical Engineering
- Medical Imaging
- Artificial Intelligence
Background:
- Continuous blood pressure (BP) monitoring is crucial for cardiovascular health management.
- Current invasive methods pose risks, necessitating non-invasive alternatives.
- Remote photoplethysmography (rPPG) offers a promising avenue for contactless BP estimation.
Purpose of the Study:
- To develop and validate a novel remote method for continuous blood pressure estimation using spatial information from facial pulse waves.
- To compare the proposed method's accuracy against conventional pulse feature analysis.
- To investigate the deep learning model's interpretability by visualizing spatial and channel-wise contributions.
Main Methods:
- Facial RGB videos were processed to extract spatial pulse-wave signals by segmenting regions of interest.
- Pulse wave contours and beat relationships were converted into time-continuous spatial signals.
- A convolutional neural network (CNN) was employed to model the relationship between spatial signals and BP.
- A dataset of continuous BP and facial videos from ten healthy volunteers was created for validation.
Main Results:
- The proposed remote method demonstrated adequate BP estimation performance, achieving a correlation coefficient of 0.85 and a mean absolute error of 5.4 mmHg.
- The method significantly outperformed conventional pulse feature-based estimation in terms of error reduction.
- Visualization revealed that spatial-wise contributions are blood pressure-dependent, while pulse contour-wise contributions reflect wave dynamics.
Conclusions:
- The proposed remote sensing method provides an accurate and non-invasive approach for continuous blood pressure monitoring.
- Deep learning-based analysis of spatial pulse-wave signals offers superior performance compared to traditional methods.
- The interpretability analysis offers insights into the physiological basis of remote BP estimation.
Related Concept Videos
Assessing Blood pressure using a doppler ultrasound
Pre-Procedural Guidelines for Doppler Ultrasound Blood Pressure Assessment:
Preparation of Equipment:
Special considerations while measuring blood pressure
Monitoring Both Arms:
Monitoring BP in both arms during the initial assessment is advisable, as the systolic value may differ by five to ten mm Hg between arms. For subsequent BP assessments, use the arm with the higher reading.
Measurement of Blood Pressure
Pre-Procedural Guidelines for Assessing Blood Pressure
Sites for measruring blood pressure
The Brachial Artery: Primary Site for Blood Pressure Measurement
Assessment of blood pressure in brachial artery(two-step method)

