Exploring the limitations of blood pressure estimation using the photoplethysmography signal

TL;DR

This study tackles the challenge of noninvasively estimating blood pressure (BP) from single-site signals, especially photoplethysmography (PPG), by proposing a calibration-based pipeline using a Siamese ResNet. It compares normalized PPG ($N$-PPG) against normalized invasive arterial BP ($N$-IABP) as inputs, using the VitalDB dataset and evaluating against AAMI and BHS standards. The key contributions are establishing an $N$-IABP–based performance upper bound, implementing a robust data pairing strategy to avoid leakage, and demonstrating that while $N$-PPG carries BP-related information, it generally cannot match $N$-IABP performance under identical conditions. The findings imply cautious optimism for PPG-based approaches in wearable BP monitoring, highlighting significant limitations and the need for rigorous validation and improved modeling to achieve clinically acceptable accuracy.

Abstract

Hypertension, a leading contributor to cardiovascular morbidity, underscores the need for accurate and continuous blood pressure (BP) monitoring. Photoplethysmography (PPG) presents a promising approach to this end. However, the precision of BP estimates derived from PPG signals has been the subject of ongoing debate, necessitating a comprehensive evaluation of their effectiveness and constraints. We developed a calibration-based Siamese ResNet model for BP estimation, using a signal input paired with a reference BP reading. We compared the use of normalized PPG (N-PPG) against the normalized Invasive Arterial Blood Pressure (N-IABP) signals as input. The N-IABP signals do not directly present systolic and diastolic values but theoretically provide a more accurate BP measure than PPG signals since it is a direct pressure sensor inside the body. Our strategy establishes a critical benchmark for PPG performance, realistically calibrating expectations for PPG's BP estimation capabilities. Nonetheless, we compared the performance of our models using different signal-filtering conditions to evaluate the impact of filtering on the results. We evaluated our method using the AAMI and the BHS standards employing the VitalDB dataset. The N-IABP signals meet with AAMI standards for both Systolic Blood Pressure (SBP) and Diastolic Blood Pressure (DBP), with errors of 1.29+-6.33mmHg for systolic pressure and 1.17+-5.78mmHg for systolic and diastolic pressure respectively for the raw N-IABP signal. In contrast, N-PPG signals, in their best setup, exhibited inferior performance than N-IABP, presenting 1.49+-11.82mmHg and 0.89+-7.27mmHg for systolic and diastolic pressure respectively. Our findings highlight the potential and limitations of employing PPG for BP estimation, showing that these signals contain information correlated to BP but may not be sufficient for predicting it accurately.

Figures (7)

• Figure 1: Illustration of different Blood Pressure Monitoring Methods.
• Figure 2: Comprehensive analysis of PPG signal components and waveform characteristics.
• Figure 3: Pairs of PPG (blue) and IABP (red) waveform samples extracted from the same patient in the VitalDB dataset for two distinct instant in time. Systolic/Diastolic values for both selected windows are labeled as (A) 120/75 mmHg, and (B) 150/90 mmHg.
• Figure 4: The proposed methodology for BP estimation.
• Figure 5: General preprocessing steps.