Share Email Print

Proceedings Paper

Acoustic sensor array extracts physiology during movement
Author(s): Michael V. Scanlon
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

An acoustic sensor attached to a person's neck can extract heart and breath sounds, as well as voice and other physiology related to their health and performance. Soldiers, firefighters, law enforcement, and rescue personnel, as well as people at home or in health care facilities, can benefit form being remotely monitored. ARLs acoustic sensor, when worn around a person's neck, picks up the carotid artery and breath sounds very well by matching the sensor's acoustic impedance to that of the body via a gel pad, while airborne noise is minimized by an impedance mismatch. Although the physiological sounds have high SNR, the acoustic sensor also responds to motion-induced artifacts that obscure the meaningful physiology. To exacerbate signal extraction, these interfering signals are usually covariant with the heart sounds, in that as a person walks faster the heart tends to beat faster, and motion noises tend to contain low frequency component similar to the heart sounds. A noise-canceling configuration developed by ARL uses two acoustic sensor on the front sides of the neck as physiology sensors, and two additional acoustic sensor on the back sides of the neck as noise references. Breath and heart sounds, which occur with near symmetry and simultaneously at the two front sensor, will correlate well. The motion noise present on all four sensor will be used to cancel the noise on the two physiology sensors. This report will compare heart rate variability derived from both the acoustic array and from ECG data taken simultaneously on a treadmill test. Acoustically derived breath rate and volume approximations will be introduced as well. A miniature 3- axis accelerometer on the same neckband provides additional noise references to validate footfall and motion activity.

Paper Details

Date Published: 23 August 2001
PDF: 12 pages
Proc. SPIE 4368, Visualization of Temporal and Spatial Data for Civilian and Defense Applications, (23 August 2001); doi: 10.1117/12.438116
Show Author Affiliations
Michael V. Scanlon, Army Research Lab. (United States)

Published in SPIE Proceedings Vol. 4368:
Visualization of Temporal and Spatial Data for Civilian and Defense Applications
Glenn O. Allgood; Glenn O. Allgood; Nickolas L. Faust, Editor(s)

© SPIE. Terms of Use
Back to Top