In-silico and ex-vivo validation of semi-automatic segmentation and patient specific implant design for the round window niche to treat inner ear disorders

Motivation The aim of this study was to validate our semi-automated segmentation and implant design approach, of the round window niche (RWN) and the round window membrane (RWM) for use in the development of patient individualized round window niche implants (RNI).

Material/Methods Two validation methods have been applied. First, an in-silico comparison of the developed semi-automatic segmentation with a previous manual segmentation based on 20 clinical cone beam computed tomography datasets of unilateral temporal bones. Two otolaryngologists, one experienced and one at the beginning of the residency, performed the semi-automated segmentation independently. Second, an ex-vivo validation of the developed software and the surgical fitting accuracy was verified in N=4 body donor implantation tests with additively manufactured RNI.

Results The volume of the RWN semi-automated segmentations of User 1 was 13 ± 12% smaller on average than the RWN segmentation of User 2. On the other hand without corrections of the manual segmentation (by example by removing bone voxels from the implant) the volume of the semi-automated RWN segmentations were 48±11 % smaller on average than the manual segmentation. Despite the differences in volume of the RWN, all additively manufactured implants based on the semi-automated segmentation were accurately fitted pressure-tight in the RWN, without room for wobbling in the RWN.

Conclusion This study presents a semi-automated approach for segmenting structures in temporal bone CBCT scans that is efficient and accurate, and not dependent on trained users.

RESPONSE–Partnership for Innovation in Implant Technology’ in the program ‘Zwanzig20– Partnership for Innovation

 

MissingLinkText

Conflict of Interest

The authors declare that they have no conflict of interest.

Publication History

Article published online:
12 May 2023