Periodic Reporting for period 1 - CIModelPLUS (Advanced Computational Model for the Development of Cochlear Implants)
Período documentado: 2017-03-01 hasta 2019-02-28
Cochlear implants (CIs) are surgically-implantable biomedical devices that bypass the sensory hair cells and directly excite the remaining fibres of the auditory nerve with electric current. They are capable of restoring a high degree of auditory perception to patients that are severe-to-profoundly deaf. Up to the year of 2012, there were more than 325,000 CI recipients all over the world, and more than 100,000 CI users in Europe, which were about 200 implanted patients per million inhabitants. However, this only accounts for 7% of all adults with hearing impairment that could benefit from a CI in Europe.
CI users usually regain the ability to understand and use spoken language with or without visual aid. However, there remains a wide variation of individual in speech recognition performance after implantation, and the ability of implant users to understand speech diminishes in the presence of background noise. Computational models of the cochlea and sensory neurons have been employed to facilitate the development of CIs via exploring various aspects of electrical stimulation. These models depend on the geometry of the reconstructed cochlea and the exact knowledge of the electrical impedance of different tissues. However, only crude estimation of the impedance of all media is available. Moreover, most existing computational models of the cochlea only adopt an ideal representation of the geometry due to limitations in structural image resolution or in computational expenses.
The aim of CIModelPLUS was to provide engineers an advanced computational model of the cochlea that facilitates the development of next-generation CI. To fulfil the aim of CIModelPLUS, the following objectives were included: 1. development of anatomically-accurate computational model of cochlea; 2. measurement of the electric current distribution; 3. generation of the advanced computational model. All milestones defined for CIModelPLUS have been reached.
CI users usually regain the ability to understand and use spoken language with or without visual aid. However, there remains a wide variation of individual in speech recognition performance after implantation, and the ability of implant users to understand speech diminishes in the presence of background noise. Computational models of the cochlea and sensory neurons have been employed to facilitate the development of CIs via exploring various aspects of electrical stimulation. These models depend on the geometry of the reconstructed cochlea and the exact knowledge of the electrical impedance of different tissues. However, only crude estimation of the impedance of all media is available. Moreover, most existing computational models of the cochlea only adopt an ideal representation of the geometry due to limitations in structural image resolution or in computational expenses.
The aim of CIModelPLUS was to provide engineers an advanced computational model of the cochlea that facilitates the development of next-generation CI. To fulfil the aim of CIModelPLUS, the following objectives were included: 1. development of anatomically-accurate computational model of cochlea; 2. measurement of the electric current distribution; 3. generation of the advanced computational model. All milestones defined for CIModelPLUS have been reached.
A three-dimensional (3D) anatomically-accurate computer model of the human cochlea with embedded CI electrode array was reconstructed based on a set of high-resolution micro-CT scans of a human temporal bone. Electric current distribution in the cochlea as a result of electrical stimulation from CI electrodes was simulated using this 3D cochlear model.
Three different tasks were performed to measure the electric current inside the cochlea, and the results were later used as a reference for optimising the computer model. Task 1 was to perform direct measurements of the electric potential within the cochlea of a temporal bone specimen by using an additional microelectrode. Task 2 was to directly visualise the electric current flow using a technique called current density imaging (CDI). It turned out that this method was not sensitive enough. We, therefore, conducted additional measurements (Task 3), where we measured electric potentials inside the cochlea of implanted patients using the implanted electrode. Subsequently, these measurements were used to optimise the electrical conductivity values of different tissues in the cochlear model. The influence of CI electrode positions on the electric current distribution in the cochlea was investigated using the optimised model.
Moreover, a semi-automated algorithm was developed to reconstruct the path of auditory nerve fibres (ANFs) inside the cochlear model by searching for the shortest path between given starting and end points. In order to implement a biophysical model on the reconstructed ANFs, we conducted a systemic review of different models in the literature. We hope that based on the review, we will be able to select or create the optimal biophysical model to be implemented within our advanced computational model of human cochlea.
Preliminary results obtained within CIModelPLUS have been presented at major conferences and invited talks within the field of medical engineering and hearing research. Final results will be published at high-impact international journals with Open Access.
Three different tasks were performed to measure the electric current inside the cochlea, and the results were later used as a reference for optimising the computer model. Task 1 was to perform direct measurements of the electric potential within the cochlea of a temporal bone specimen by using an additional microelectrode. Task 2 was to directly visualise the electric current flow using a technique called current density imaging (CDI). It turned out that this method was not sensitive enough. We, therefore, conducted additional measurements (Task 3), where we measured electric potentials inside the cochlea of implanted patients using the implanted electrode. Subsequently, these measurements were used to optimise the electrical conductivity values of different tissues in the cochlear model. The influence of CI electrode positions on the electric current distribution in the cochlea was investigated using the optimised model.
Moreover, a semi-automated algorithm was developed to reconstruct the path of auditory nerve fibres (ANFs) inside the cochlear model by searching for the shortest path between given starting and end points. In order to implement a biophysical model on the reconstructed ANFs, we conducted a systemic review of different models in the literature. We hope that based on the review, we will be able to select or create the optimal biophysical model to be implemented within our advanced computational model of human cochlea.
Preliminary results obtained within CIModelPLUS have been presented at major conferences and invited talks within the field of medical engineering and hearing research. Final results will be published at high-impact international journals with Open Access.
CIModelPLUS, by developing an advanced computational model of the cochlea, serves to facilitate the development of next-generation CIs. It will therefore fortify the European prominence and expertise and defend the leadership of European Union (EU) in the research field of CI development. In addition, CIModelPLUS aims to tackle EU societal challenge Health, Demographic Change and Wellbeing by improving our understanding of the CI working mechanism, and by testing and demonstrating new tool for CI development. It will contribute to improving the life quality of more than 100,000 current CI users and other CI candidates in Europe, as well as that of their families and the communities that they reside in.