Acoustic reflections during rhinometry: spatial resolution and sound loss.

The accuracy of the acoustic reflections method for the evaluation of human nasal airway geometry is determined by the physical limitations of the technique and also by the in vivo deviations from the assumptions of the technique. The present study 1) examines the sound loss caused by nonrigidity of the nasal mucosa and viscous loss caused by complex geometry and its influence on the estimation of the acoustic area-distance function; 2) examines the optimal relation between sampling frequency and low-pass filtering, and 3) evaluates advantages of breathing He-O2 during the measurements on accuracy. Measurements made in eight plastic models, with cavities exactly identical to the "living" nasal cavities, revealed only minor effects of nonrigidity of the nasal mucosa. This was confirmed by an electrical analog model, based on laser vibrometry admittance measurements of the nasal mucosa, which indicated that the error in the acoustic measurements caused by wall motion is insignificant. The complex geometry of the nasal cavity per se (i.e., departure from circular) showed no significant effects on the measurements. Low-pass filtering of the signal is necessary to cut off cross modes arising in the nasal cavity. Computer simulations and measurements in models showed that the sampling frequency should be approximately four times the low-pass filtering frequency (i.e., twice the Nyquist frequency) to avoid influence on the result. No advantage was found for the the use of He-O2 vs. air in the nasal cavity.