A computer model can predict and improve the performance and response of cochlear implants

For advances in neuroscience, the biomedical engineering techniques that have been most successful are deep brain stimulation and cochlear implants. Cochlear implantation is a surgical procedure that aims for the patient to regain his/her sense of hearing through electrical stimulation directly to the cochlea of the inner ear.

Despite surgery restoring hearing after the loss of sensitivity, the exact degree to which this occurs is highly variable for each patient. Among the various factors at play are the design of the implant, its placement and the stimulation protocol. However, there is a lack of understanding as to how each of these factors affects neural propagation and excitement whilst the implant is functioning.

A study published in June in the journal Molecular Neurobiology, coordinated by González Ballester, ICREA researcher of the Department of Information and Communication Technologies (DTIC) and head of the SiMBioSys (Simulation, Imaging and Modelling for Biomedical Systems) Research Group at UPF, describes how the performance and response to surgical cochlear implantation can be predicted through computer models that can be redesigned according to the results of the operation in order to provide patients with more efficient stimulation regimens.

The study has involved team members Mario Ceresa and Nerea Mangado, and Russell J. Andrews, a scientist of NASA's Intelligent Systems in Nanotechnology (USA) group.

Personalized medicine in the field of cochlear implants

As González Ballester commented, "in this work we show how highly accurate computer models specific to each patient provide essential information to improve the interface between the electrode and the nervous system".

The researchers worked with two groups of patients, one presenting intact nerve fibres and the other with degenerated nerve fibres. Then, using the computer model they have managed to predict the response to electrical stimulation of each patient and redesign the stimulation protocol to use the optimal parameters.

"We believe that our work has the potential to help improve the quality of life of patients in two ways. First, helping to design better implants and stimulation protocol, and secondly, adjusting the technique according to the anatomy of each patient before surgery, with the help of medical images", González Ballester stresses. "This will enable customized planning for surgery and postoperative follow-up for each patient within a scenario of personalized medicine", he added.

This study is part of the European project Hear-EU, coordinated by Miguel A. González Ballester. The main goal of this project is to develop new computer systems based on high resolution microCT images (computerized microtomography or 3D X-ray imaging), and predictive mathematical models to help surgery and the design of new implants.

Reference work:

Mario Ceresa, Nerea Mangado, Russell J. Andrews, Miguel A. Gonzalez Ballester, (2015), " Computational Models for Predicting Outcomes of Neuroprosthesis Implantation: the Case of Cochlear Implants", Molecular Neurobiolo gy, DOI 10.1007/s12035-015-9257-4.