Dependence of Cochlear Duct Length Measurement on the Resolution of the Imaging Dataset

Bjoern Spahn,Lukas Ilgen,Tilmann Neun,Franz-Tassilo Müller-Graff,Philipp Schendzielorz,Rudolf Hagen,Daniel Althoff,Simon Zabler,Kristen Rak
DOI: https://doi.org/10.1097/mao.0000000000004088
2024-01-10
Otology & Neurotology
Abstract:Hypothesis Measurements of the cochlear duct length (CDL) are dependent on the resolution of the imaging dataset. Background Previous research has shown highly precise cochlear measurements using 3D-curved multiplanar reconstruction (MPR) and flat-panel volume computed tomography (fpVCT). Thus far, however, there has been no systematic evaluation of the imaging dataset resolution required for optimal CDL measurement. Therefore, the aim of this study was to evaluate the dependence of CDL measurement on the resolution of the imaging dataset to establish a benchmark for future CDL measurements. Methods fpVCT scans of 10 human petrous bone specimens were performed. CDL was measured using 3D-curved MPR with secondary reconstruction of the fpVCT scans (fpVCT SECO ) and increasing resolution from 466 to 99 μm. In addition, intraobserver variability was evaluated. A best-fit function for calculation of the CDL was developed to provide a valid tool when there are no measurements done with high-resolution imaging datasets. Results Comparison of different imaging resolution settings showed significant differences for CDL measurement in most of the tested groups ( p < 0.05), except for the two groups with the highest resolution. Imaging datasets with a resolution lower than 200 μm showed lower intraobserver variability than the other resolution settings, although there were no clinically unacceptable errors with respect to the Bland-Altman plots. The developed best-fit function showed high accuracy for CDL calculation using resolution imaging datasets of 300 μm or lower. Conclusion 3D-curved MPR in fpVCT with a resolution of the imaging dataset of 200 μm or higher revealed the most precise CDL measurement. There was no benefit of using a resolution higher than 200 μm with regard to the accuracy of the CDL measurement.
clinical neurology,otorhinolaryngology
What problem does this paper attempt to address?