A cochlear scaling model for accurate anatomy evaluation and frequency allocation in cochlear implantation

Daniel Schurzig*, Markus Pietsch, Peter Erfurt, Max E. Timm, Thomas Lenarz, Andrej Kral

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

14 Citations (Scopus)

Abstract

The human cochlea has a highly individual microanatomy. Cochlear implantation therefore requires an evaluation of the individual cochlear anatomy to reduce surgical risk of implantation trauma. However, in-vivo cochlear imaging is limited in resolution. To overcome this issue, cochlear models based on exact anatomical data have been developed. These models can be fitted to the limited parameters available from clinical imaging to provide a prediction of the precise cochlear microanatomy. Recently, models have become available with improved precision that additionally allow predicting the 3D form of an individual cochlea. The present study has further improved the precision of modelling by incorporating microscopic details of a large set of 108 human cochleae from corrosion casts. The new model provides a more flexible geometric shape that can better predict local variations like vertical dips and jumps and provides an approximation of frequency allocation in the cochlea. The outcome of this and five other models have been quantified (validated) on an independent set of 20 µCTs of human cochleae. The new model outperformed previous models and is freely available for download and use.

Original languageEnglish
Article number108166
Pages (from-to)1-11
Number of pages11
JournalHearing Research
Volume403
Early online date5 Jan 2021
DOIs
Publication statusPublished - 1 Apr 2021

Keywords

  • cochlear duct length
  • Cochlear implantation
  • individualization
  • insertion angle prediction
  • surgical planning

Fingerprint

Dive into the research topics of 'A cochlear scaling model for accurate anatomy evaluation and frequency allocation in cochlear implantation'. Together they form a unique fingerprint.

Cite this