Ramos-de-Miguel Ángel, Escobar José M, Greiner David, Ramos-Macías Ángel
University Institute of Intelligent Systems and Numerical Applications in Engineering (SIANI), University of Las Palmas de Gran Canaria, Las Palmas, Spain.
Otolaryngology Head and Neck Surgery, University and Children's Hospital Insular of Las Palmas, Las Palmas, Spain.
Int J Numer Method Biomed Eng. 2018 Apr 6:e2992. doi: 10.1002/cnm.2992.
This paper presents a new procedure to design optimal electrodes for cochlear implants. The main objective of this study is to find a set of electrode designs that maximize the focalization and minimize the power consumption simultaneously. To achieve that, a criterion to measure the ability of focalization of an electrode is proposed. It is presented a procedure to determine (1) the electrical potential induced by an electrode by solving the Laplace equation through the finite element method; (2) the response of a neuron to an applied field using NEURON, a compartmentalized cell model; (3) the optimization to find the best electrode designs according to power consumption and focalization by 2 evolutionary multiobjective methods based on the non-dominated sorting genetic algorithm II: a straight multiobjective approach and a seeded multiobjective approach. An electrode design formed by 2 conductive rings with a possible difference of potential between them is proposed. It is analyzed that the response of the neuron is determined by the shape and the difference of the potential between the electrode rings. Our procedure successfully achieves a nondominated set of optimum electrode designs improving a standard electrode in both objectives, as designs with better focalization allow to include extra electrodes in the cochlear implant, and designs with lower power consumption extend the length of the battery.
本文提出了一种为人工耳蜗设计最佳电极的新方法。本研究的主要目标是找到一组电极设计，使其聚焦效果最大化，同时功耗最小化。为实现这一目标，提出了一种衡量电极聚焦能力的标准。介绍了一种确定方法：(1) 通过有限元法求解拉普拉斯方程来确定电极感应的电势；(2) 使用神经元（一种 compartmentalized 细胞模型）来确定神经元对施加电场的响应；(3) 通过基于非支配排序遗传算法 II 的两种进化多目标方法，根据功耗和聚焦情况进行优化以找到最佳电极设计：一种直接多目标方法和一种种子多目标方法。提出了一种由两个导电环组成的电极设计，它们之间可能存在电势差。分析表明，神经元的响应由电极环的形状和电势差决定。我们的方法成功地获得了一组非支配的最佳电极设计，在两个目标方面都优于标准电极，因为聚焦效果更好的设计允许在人工耳蜗中增加额外电极，而功耗更低的设计可延长电池续航时间。
