Journal Club: Finite element modelling of the human skull to understand how head securing might impact bone conduction.

on

Today's journal article

Lim J, Shin HS, Dobrev I, Niermann J, Röösli C, Yoon YJ, Kim N. Exploring the effect of cadaver head fixation in bone conduction hearing: Insights from finite element modeling. 

Why I picked this article

This research used a combination of experiments with a secured human skull and finite element modelling to understand how vibration is transmitted across the bone, which relates to our bone-conduction hearing. 

Our sense of hearing has two underlying types of hearing - air-conduction hearing and bone-conduction hearing. When we consider sounds travelling as the vibration of the air, hitting and vibrating the eardrum and causing vibration of the ossicles, this is air-conduction hearing. Bone conduction hearing bypasses the ear canal and middle ear; instead, the sound is transmitted through the bone of the skull. Bone-conduction hearing is particularly important for hearing our internal sounds, like our own voice. When we hear a recording of our own voice, it always sounds very different to how we normally hear it. This is because of the bone conduction hearing, which normally plays a large role in hearing our own voice. 

Understanding bone conduction hearing has led to new technologies like bone-conduction headphones, which you can wear without blocking your ear canal, and bone-conduction hearing aids and bone-conduction hearing assessment diagnostic tools. In this research, to understand bone conduction better, researchers have built a finite-element model of a human skull and simulated bone conduction hearing to compare with data from existing literature. 

Some of the research findings

Finite element model

  • Head model components: soft tissue, cortical bone, cancellous bone, cerebrospinal fluid, brain, cartilage and eyes.
  • Auditory periphery model: middle ear, cochlear outer bone, cochlear fluid, basilar membrane
  • Simulation conducted using ACTRAN (Commercial package from Hexagon.com?)
  • Frequencies for simulation: 0.1 - 10kHz with 0.1kHz increment

Head fixation

  • Three types of methods to secure a skull used in literature were used. 
  • Cadaver head resting on an anti-vibration pad
  • Four pins stablising the cadaver head
  • a Mayfield head clamp to secure the cadaver head (with Young's modulus, or elasticity, of ~100GPa). 
Simulation outputs;
  • pulmonary acceleration 
  • transcranial attenuation (TA)
  • Validation of the model

Figure 1(a). Finite element model of the skull. Lim et al. 2025

Findings:
  • Simulation of transcranial attenuation (how much of bone-conducted sound is attenuated by the skull) aligned very closely with the experimental data reported in the literature. 
  • The pattern of transcranial attenuation profile across frequencies tested differed depending on the head fixation modalities; attenuation was more at higher frequencies (3kHz and above) with an anti-vibration pad placed under the skull. Attenuation was more profound in 0.5-1kHz frequency range with four-pin fixation. 
  • Fixation strength and height did not impact the attenuation profile very much. 
  • Having neck support had almost no impact on the attenuation. 

Haruna's takeaway

The frequency-dependency of the attenuation by the bone was interesting. The impact from different ways of securing the skull was large, too, which I guess makes a lot of sense. I wonder how this may affect how we should be collecting functional assessment data using a bone conduction probe. For example, we conduct animal experiments using sheep as the large animal model and test bone-conduction hearing. Maybe there should be some consideration for how we are securing the head. 

I would also really like us to get into modelling and simulation. So I will be reading and sharing more simulation-related work in the upcoming journal club!  

 ------- 

This is Haruna's 29/100 of the 100-day challenge to post a science blog article every day! I love inner ear biology & cochlear physiology.