A new validation range for temporal bone experiments

Link to publication (open source)

Measurements of the middle ear transfer function (METF) in human temporal bone specimens are fundamental to the research of middle ear mechanics. These experiments are essential in the development of new instruments, surgical techniques, and prostheses as well as active hearing implants. When drawing conclusions from these measurements, it is imperative that the samples used are representative of the target population. Due to anatomical variation between subjects, this requires a standardized definition of "normal" sound transmission. The METF is the appropriate metric for this purpose. An important step towards a standardized range was undertaken by Rosowski et al. and the corresponding standard was widely adapted. However, the suitability of said range for the intended purpose has recently been called into question. Reliable statistics with a unified database are still lacking.

A new paper published in the open source journal Scientific Reports (Springer Nature) present a statistical analysis of METF measurements in 478 temporal bones from 5 different international research groups. From these, statistical parameters are derived which are attributed to non-pathological sound transfer. This includes the mean, 95% confidence interval of the mean, and the 95% population-proportion tolerance interval. Differences between research groups and measurement methods are evaluated using Welsh's t-tests and linear mixed model analysis. The publication further details how the different variables in the experimental setup and procedure affect the measurements and the statistical parameters of a sample group. The multi-center data is then grouped for a clearly defined measurement method and the target statistics are calculated

The final result of the publication are statistical intervals which may be used as an objective range for the validation of individual temporal bone specimens.

Extensive supplementary materials are published along with the results, so that the current database can be expanded and the accuracy of the ranges improved as more METF measurements become available.