Evaluation of one-point calibration for cuffless BP wearable sensor devices: Stiffness Index

Hongqiang Xu, Fatemeh Heydari, Asanka Rathnayaka, Fan Wu, Mehmet Rasit Yuce

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Blood Pressure (BP) is one of the most important vital signs for the early diagnosis of cardiovascular diseases (CVDs). This paper presents a non-invasive continuous BP measurement system using one-point calibration. The conventional continuous BP measurement in wearable sensor devices uses Pulse Arrival Time (PAT) or Pulse Transient Time (PTT), which requires multiple measurements for calibration. Additionally, users need to increase their BP through exercising for calibration. In this study, we developed a novel algorithm that only requires one steady BP measurement for calibration. This calibration method is based on the population calibration parameters and their artery stiffness information from Photoplethysmography (PPG) signal. The single calibration method utilizes a two-stage linear regression approach based on people’s arterial stiffness index (SI) for wearable devices. Methods: PPG and Electrocardiogram (ECG) sensors are attached to the hands to measure analog signals. Simultaneously, a cuff device is attached to the right arm as a reference. After raw PPG and ECG signals are processed with filters, landmarks can be successfully extracted using our proposed algorithm. We have tested the data with different body postures (sitting, standing, and supine) as well as varying levels of exercise intensity. We collected and analyzed data from 41 individuals, including 30 males and 11 females for this study. Finally, individual and population linear regression methods are applied to the data to reduce the calibration process. The results of this study give an average mean absolute error (MAE) of 3.832 and an average R2 of 0.8034 for systolic blood pressure (SBP). 87.33% of data are within the accuracy of ±10 mmHg.

Original languageEnglish
Pages (from-to)11374-11385
Number of pages12
JournalIEEE Sensors Journal
Volume24
Issue number7
DOIs
Publication statusPublished - 1 Apr 2024

Keywords

  • Biomedical monitoring
  • Blood pressure
  • BP calibration
  • Calibration
  • Cuffless BP estiamtion
  • ECG
  • Electrocardiography
  • Electrodes
  • Estimation
  • PPG
  • Sensors
  • Stiffness Index

Cite this