Smart insoles are a tool with great potential for monitoring physiological parameters, including heart rate (HR), whose measurement on the sole of the foot remains a challenge. In this work, a new system for beat-to-beat HR detection based on the measurement of the ballistocardiogram (BCG) with a piezoelectric sensor embedded in a commercial insole is presented. Ballistocardiography is a noninvasive method for measuring whole-body recoil forces in reaction to cardiac ejection of blood through the vascular tree at each heartbeat, thus eliminating the need to detect any specific artery or to establish direct contact with the skin. The validation of the proposed system was done with 25 subjects: BCG and an ECG lead I were acquired simultaneously for 2 min while standing or sitting with no instructions to maintain silence or refrain from moving. A signal segmentation process was performed to discard segments contaminated by motion artifacts, achieving a mean coverage factor above 95% in both positions. For beat-to-beat HR detection from the BCG, an algorithm based on the Continuous Wavelet Transform was used, which yielded a mean of TP (true positives) detected of 82.7% in standing position and 91.8% in seated position. To assess the agreement between the HR values obtained from BCG and ECG, a Bland-Altman plot was used to compare them. The 95% confidence intervals in beats per minute (bpm) were ±3.64 bpm and ±3.21 bpm in the standing and seated position respectively, meeting the accuracy limits recommended by the AAMI standard for HR meters.
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes,creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
