The effects of varying tympanic-membrane material properties on human middle-ear sound transmission in a three-dimensional finite-element model.

An anatomically based three-dimensional finite-element human middle-ear (ME) model is used to test the sensitivity of ME sound transmission to tympanic-membrane (TM) material properties. The baseline properties produce responses comparable to published measurements of ear-canal input impedance and power reflectance, stapes velocity normalized by ear-canal pressure (P), and middle-ear pressure gain (MEG), i.e., cochlear-vestibule pressure (P) normalized by P. The mass, Young's modulus (E), and shear modulus (G) of the TM are varied, independently and in combination, over a wide range of values, with soft and bony TM-annulus boundary conditions. MEG is recomputed and plotted for each case, along with summaries of the magnitude and group-delay deviations from the baseline over low (below 0.75 kHz), mid (0.75-5 kHz), and high (above 5 kHz) frequencies. The MEG magnitude varies inversely with increasing TM mass at high frequencies. Increasing E boosts high frequencies and attenuates low and mid frequencies, especially with a bony TM annulus and when G varies in proportion to E, as for an isotropic material. Increasing G on its own attenuates low and mid frequencies and boosts high frequencies. The sensitivity of MEG to TM material properties has implications for model development and the interpretation of experimental observations.