Abstract:
Imaging the cochlea in vivo is a challenging task. Regardless of the quality of images obtained from modern clinical imaging techniques, the internal structures of the cochlea remain mainly obscured. Electrical impedance tomography (EIT) is a safe, low-cost alternative medical imaging technique, with applications in a variety of clinical scenarios. In this study, EIT is investigated as an alternative method to image and extract the centre of gravity of the modiolus (CoG) in vivo. This information can be used to augment present postoperative medical imaging techniques to investigate the cochlea.
The cochlear implant (CI) electrode array was used in conjunction with the EIT software Electrical Impedance Tomography and Diffuse Optical Reconstruction Software (EIDORS) to develop a CI-EIT system. A methodology was successfully developed to import models of varying complexity, along with the electrode contacts and domains of arbitrary conductivity, from COMSOL to EIDORS. An adapted stimulation protocol compliant with CI specifications was also developed and employed, in conjunction with the imported models, for EIT image reconstruction using CI electrode arrays.
The CI-EIT system was simulated by modelling user-specific electrode array trajectories within a simple conductive medium containing an inhomogeneity representing the modiolus. The method included an adapted adjacent stimulation protocol for data collection. For the image reconstruction, NOSER and Tikhonov priors were considered. A parameter analysis was conducted to find the most robust combination of image priors and hyperparameters for this application. The CI-EIT system methodology was validated at different noise levels for four electrode array trajectories. Comparing the NOSER and Tikhonov priors, it was observed that the NOSER prior exhibits superior centre of gravity localisation performance in cochlear implant EIT image reconstruction for different noise levels and user-dependent variability in electrode array trajectories. Image reconstruction, using a NOSER prior at a hyperparameter value of approximately 0.001, resulted in an average centre of gravity localisation error of less than 4 % for all electrode array trajectories using difference imaging, and less than 5.5 % using absolute imaging. Using the same optimal image reconstruction parameters, an average conductivity estimation error ranging between 10 % to 50 % for all electrode array trajectories was observed.
After the establishment of the most robust image reconstruction parameters, four new anatomically realistic mesh geometries with unique electrode trajectories were implemented. Additionally, a simple mesh containing only the four electrode trajectories in a volume, and a duplicate of the latter with mesh refinement in the region of interest was implemented. The effect of the complexity of the mesh on the CoG localisation was investigated. The CoG localisation using the anatomically realistic mesh geometry provided marginally more accurate results than the simple mesh implementations, with a maximum of only 4 % improvement. However, the simpler mesh geometries decreased the computational time for reconstruction to less than 30 % of the time required for solving the anatomically realistic mesh for difference image, and to less than 25 % of the time required for solving the absolute image using the anatomically realistic mesh. The ability of the proposed CI-EIT system to reconstruct typically obscured internal structures of the cochlea, such as the modiolar wall trajecory, was investigated. A discrete Frechet distance analysis revealed that the trajectory of the automatically extracted modiolar wall from the EIT reconstruction is similar to the theoretical trajectory. A Procrustes analysis also revealed that the Bezier curves representing the modiolar wall are similar in shape among the three mesh complexities.
It was concluded that the proposed CI-EIT system to image and extract geometric properties of the modiolus is a viable proposition. However, more research and a multidisciplinary approach are required for the proposed CI-EIT system to become a clinical reality.
