Automated 3D Inner Ear Assessment for Optimizing Animal Models of Noise-Induced Hearing Loss

ObjectiveTo facilitate the selection of optimal animal models for noise-induced hearing loss (NIHL) research, this study aims to develop an accurate, automatic centerline-based method to construct three-dimensional (3D) inner ear models in various mammals and assess their morphological similarities to humans. MethodsThree adult mice, three guinea pigs, three mini pigs, and one human temporal bone were scanned using Micro-CT. After segmentation, 3D inner ear models were reconstructed. A novel centerline method was established to automatically estimate 3D properties and calculate length, volume, and angle parameters across species. ResultsDetailed 3D models of the cochlea, vestibule, and semicircular canals were successfully constructed. Mean lengths and volumes of the cochlea and semicircular canals generally increased with body size, with mini pigs demonstrating the closest proximity to human data. Notably, numerical assessments showed that the semicircular canals in both mice and mini pigs closely resembled human structures. ConclusionAutomatic centerline-based measurement provides an effective tool for assessing 3D inner ear morphology. By confirming the high morphological similarity between mini pigs and humans, this study provides a critical theoretical basis for the mechanical analysis of noise-induced damage and the selection of biomimetic animal models for NIHL studies.