Acoustic rhinometry: a study of transient and continuous noise techniques with nasal models

The objective of this study is to compare the properties of two of the most frequently used acoustic rhinometers: the EcoVision (Hood Laboratories, USA) using the transient technique, and the Rhin2100 (RhinoMetrics, Denmark) using the continuous wide-band technique. In the wide-band rhinometer (Rhin2100), the transient analog signals of traditional rhinometers (EcoVision), are replaced by a digitally produced continuous wide-band noise signal. Tubular models and a plastic model produced by stereolithography (SLA), representing the true replicate of the nasal anatomy, were used to compare the accuracy of the two rhinometers. The effect of increasing angling (0-50 degrees) between the sound wave tube and the cavity was evaluated in a tubular model. The curves obtained with the two rhinometers showed close similarity, and the acoustically derived volumes correlated well with the volumes of tubular (% error < 4%) as well as the complex nasal model (% error < 10.5%). Both rhinometers underestimated the minimum cross-sectional area (MCA) of the complex nasal model (mean % error complex model: Rhin2100 = -7.6%, EcoVision = -13%). The effect of increasing the angle between the nose adapter and the tubular models was small for both rhinometers (CV < 3% for MCA and CV < 1% for volumes). The similar, and in general, high accuracy of the two rhinometers evaluated, particularly in the complicated geometry of the SLA model, is an indication of the reliability of both. The small effect of changing the angle between the nose adapter and the models was unexpected and very encouraging. Nevertheless, some minor differences in performance and capabilities of the two rhinometers might influence interpretation and comparison of results. Further comparisons in a clinical setting are under current investigation.