Semi-automatic algorithm to build finite element numerical models of the human hearing system from Micro-CT data

Area of Science:

• Biomechanics
• Medical Imaging
• Computational Modeling

Background:

• Finite Element (FE) modeling is complex for the hearing system due to geometric uncertainties.
• Existing models are often generic, limiting patient-specific validation and application.
• Developing patient-oriented FE models is time-consuming and computationally expensive.

Purpose of the Study:

• To automate the process of building patient-specific Finite Element models of the hearing system.
• To reduce the computational cost and development time for hearing system FE models.
• To enable patient-oriented modeling strategies for the tympanic membrane, ossicular chain, and cavities.

Main Methods:

• Utilized geometrical information from CT scans for model construction.
• Developed a procedure for importing CT data and building FE models.
• Integrated tasks like contour identification and model decimation for semi-automation.

Main Results:

• Reduced hearing system FE model development time from 4 weeks to approximately 3 days.
• Successfully automated the coupling of anatomical elements (tympanic membrane, ossicular chain, cavities).
• Achieved a semi-automated process for generating patient-oriented FE models.

Conclusions:

• The developed semi-automated methodology significantly accelerates the creation of patient-specific hearing system FE models.
• This approach facilitates more personalized and validated biomechanical simulations of the human ear.
• Further automation is possible, paving the way for efficient patient-oriented computational audiology.