3D structure of human auditory nerve and its impact on exitation with cochlear implants

Cochlea has a sophisticated 3D spiral shape consisting of three fluid-filled chambers and is responsible for sound perception. Sound is converted to electrical signals by sensory hair cells. The signals are transferred to brainstem through type I myelinated auditory nerve fibers (ANFs). Disruption in spiking pattern causes problems in speech understanding, particularly in hearing deficit cases. ANFs degeneration causes permanent hearing loss. However, some ANFs survive for years, which can be excited by a cochlear implant (CI), a surgically implanted device that stimulates survived ANFs extracellularly by bypassing damaged regions. CIs have critical limitations, leading to CI users' dissatisfactions with understanding speech in noisy environments and having poor music perception. Computational studies can investigate aspects to overcome limitations and improve CIs functionality, which is challenging to examine in humans. This dissertation uses a detailed finite element (FE) human cochlea model, realistic 3D helical and spiral human ANFs pathways, and FE models of CIs that are essential for a reliable computational investigation to study the temporal fine structure of the transmitted neural code tonotopic behavior of electrically stimulated ANFs. The first aim was to study degeneration patterns of human ANFs based on hearing loss levels. Synaptic excitation and the excitation properties of electrical potential distributions induced by two CIs were analyzed. The main findings indicated: (i) the unimodal distribution of peripheral axon diameters (PD) in the healthy case becomes multimodal for hearing loss cases, including numerous thin fibers. (ii) Postsynaptic currents excite thin fibers easier and earlier than healthy and larger fibers. However, this advantage causes conduction velocity to decrease proportionally with the diameter and increased spike latency and jitter. (iv) CIs cause several action potential (AP) sites in the peripheral axon, soma region, and central axon. (v) Thresholds of investigated ANFs and co-stimulation of degenerated ANFs in other frequency regions depend on the electrode position, cochlear turns, and 3D ANF pathway.Polarity sensitivity is a promising way to evaluate the neural status of ANFs in CI users. The second aim was to investigate polarity behavior using symmetric and asymmetric pulse shapes, monopolar and multipolar CI stimulation strategies, and perimodiolar and lateral CI array systems. The main findings are: (i) AP initiation sites occurred mainly in the peripheral site in lateral array regardless of stimulation strategies, pulse polarities, pulse shapes, cochlear turns, and ANF degeneration levels. However, in perimodiolar array, AP sites varied between peripheral and central processes, depending on stimulation strategies, pulse shapes, and polarities. (ii) In perimodiolar array, clusters formed in thresholds based on cochlear turns and degeneration levels for multipolar strategies only when asymmetric pulses were applied. (iii) In perimodiolar array, a declining trend in polarity ratio (anodic threshold/cathodic threshold) with multipolar strategies was observed between intact or slight degenerated cases and more severe degenerated cases, whereas in lateral array, cathodic sensitivity was noticed for intact and less degenerated cases and anodic sensitivity for cases with high degeneration degrees. The results suggest that a combination of asymmetric pulse shapes, applying multipolar stimulation strategies, and considering modiolus wall distance, allows us to distinguish the degeneration patterns of ANFs across the cochlea.The third objective was to analyze ANFs excitation based on scala tympani (ST) versus scala vestibuli (SV) CI insertion. Due to some circumstances, such as ossification and obstruction of ST, the CI insertion can be done in the SV. For this investigation, several ANFs with similar distances to the center of electrodes were selected in ST and SV by considering the different ANF degeneration levels. The main results demonstrate that excitation thresholds of ST and SV stimulating electrodes were cathodic sensitive in the intact case, and the AP occurs in peripheral processes regardless of electrode positions. In contrast, ANF excitation behavior changed in the degenerated cases, and AP differed depending on polarity and electrode positions. 3D ANF pathways significantly impact the excitation threshold and AP sites.19