| Abstrakt: |
This methodological study examined the predictive strength associated with modeling respiratory volumes from chest-wall movements during selected ventilatory and speech activities. A linearized magnetometry system transduced the anterior-posterior diameters of the rib cage and abdomen, supplying kinematic data that were used to estimate respiratory volumes. Kinematic and airflow measures were acquired during (a) tidal ventilation, (b) vital capacity maneuvers, (c) Rainbow Passage reading at customary loudness, (d) Rainbow Passage reading at twice-customary loudness, (e) extemporaneous speech, and (f) /α/ prolongation. Multiple-regression statistics were applied to the body surface and integrated airflow data to obtain (a) intercepts, (b) volume-motion coefficients for the rib cage and abdomen, and (c) coefficients of determination. Volumes estimated by applying regression-derived intercepts and volume-motion coefficients to the respiratory kinematic data were then compared to integrated airflow signals. Two magnetometer-based, volume-estimation strategies were contrasted for the speech tasks, one based on volume-motion coefficients derived from that particular speech activity and one based on volume-motion coefficients derived from tidal ventilation. Regression-derived intercepts and volume-motion coefficients for the abdomen varied significantly across subjects but not tasks. Volume-motion coefficients for the rib cage varied significantly across both subjects and tasks. Coefficients of determination for the magnetometer-based volume estimates were affected significantly by subjects and tasks but not by volume-estimation strategies. The calibration, use, and potential limitation of respiratory kinematic devices for speech research are discussed. |
