Statistical shape modeling of the human inner ear through micro-computed tomography imaging.

Area of Science:

• Anatomy
• Medical Imaging
• Otology

Background:

• The human inner ear (IE) exhibits significant morphological variability influencing function and pathology.
• Previous studies were limited by low-resolution imaging and small sample sizes.
• Understanding IE structural variability is crucial for diagnosing malformations and optimizing treatments.

Purpose of the Study:

• To characterize the bony morphology of the healthy human IE using the largest dataset of high-resolution micro-CT images.
• To identify key regions contributing to IE morphological variability.
• To investigate sex-based differences in IE morphology and develop predictive models.

Main Methods:

• Micro-computed tomography (micro-CT) imaging of 54 cadaveric temporal bone specimens.
• Semi-automatic segmentation and 3D surface mesh model creation for analysis.
• Development of statistical shape models (SSMs) for the IE, cochlea, and vestibular system, including sex- and side-based subgroups.

Main Results:

• Established normative ranges for IE dimensions, consistent with prior reports.
• Identified significant sex-based morphological differences and strong linear relationships in dimensions and volumes.
• SSMs highlighted semicircular canals, cochlear basal turn, and hook regions as primary sources of variability, with distinct patterns between sexes.

Conclusions:

• Detailed 3D characterization of healthy human IE morphological variability was achieved.
• Findings provide a foundation for assessing IE malformations and optimizing otologic interventions.
• Developed accurate multivariate models for predicting IE volumes from clinical scan data.