Image-based frequency reallocation for optimizing cochlear implant programming

描述(由申请人提供)：本研究的目标是开发和评估一种确定植入人工耳蜗物(CI)位置的方法的临床实用性。 与刺激靶点(螺旋神经节(SG)神经)相关的电极，用于CI调谐辅助。人们普遍认为，对于特定的声音，通过刺激自然对应于该声音频谱的神经可以达到最佳的听力恢复结果。然而，由于几个技术限制，这目前是不可能的。其中一个问题是，植入电极的位置是未知的。因此，听觉学家仅根据患者的反应来调整分配给每个电极的信号特征。大多数潜在的可调整参数保留为CI制造商确定的默认设置。由于这种一刀切的方法，调谐过程可能不会导致所有受试者的最佳听力恢复。每个电极的位置是可变的 深度和双极周距离。电极深度差异导致频移伪影，即每个电极刺激与检测到的声音的频率不对应的神经。到SG的距离越大，每个电极的电流分布就越大，从而降低了频谱分辨率，即每个电极刺激对应于广泛频率范围的许多神经。在未来的工作中，我们希望测试一系列先进的调谐技术，这些技术依赖于知道植入电极相对于声调映射的SG神经的位置。这些技术有可能改善现有CI所取得的听力结果。然而，还没有开发出一种技术来准确评估电极在体内相对于刺激目标的位置。在这项研究中，我们将开发这项技术，并测试一个简单的调整方案，以评估其临床应用。如果成功，我们将开发出一种易于使用的软件包，有可能提高现有人工耳蜗术的有效性，并改善植入者的听力恢复质量。为了识别电极位置，现有技术 为了识别术后CT中的耳蜗线，我们将把超前电极阵列扩展到其他类型的电极。为了识别刺激目标(SG)，将探索先进的主动形状建模技术。这些都是强大的图像处理方法，可以识别图像中的结构，同时限制结果的形状，使其与典型的解剖变异保持一致。SG区域将使用描述该频率关系的已知启发式来进行音调映射。然后，将在植入者身上测试一个简单的调整方案，以评估这种方法的实用性。调谐参数将包括确定哪些电极接通和断开，确定每个电极的相对功率，以及确定电极频率分配表。积极的结果将证明，这些技术将导致更有效的植入调整和更好的听力恢复。 与公共健康相关：拟议的项目涉及探索算法方法，这些方法将导致人工耳蜗电极阵列的性能优化。因此，该项目的结果可能会改善人工耳蜗者的听力恢复质量，并提高调谐过程的效率。

DESCRIPTION (provided by applicant): The goals of this research are to develop and assess the clinical utility of an approach for determining the position of implanted cochlear implant (CI) electrodes relative to stimulation targets (the nerves of the Spiral Ganglion (SG)) for CI tuning assistance. It is widely believed that the best hearing restoration outcome can be achieved by stimulating, for a particular sound, the nerves that naturally correspond to the spectrum of that sound. However, this is not currently possible due to several technical limitations. One such issue is that the positions of the implanted electrodes are unknown. Thus, the audiologist adjusts the signal characteristics assigned to each electrode based solely on patient response. The majority of potentially adjustable parameters are left at the default settings determined by the CI manufacturer. Because of this one-size-fits-all approach, the tuning process may not result in optimal hearing restoration for all recipients. Each electrode is positioned at variable depths and perimodiolar distances. Electrode depth discrepancies result in a frequency shift artifact, i.e., each electrode stimulates nerves that do not correspond to the frequencies of the detected sound. A larger distance to the SG leads to wider current spread from each electrode, decreasing the spectral resolution, i.e., each electrode stimulates many nerves corresponding to a wide range of frequencies. In future work, we would like to test a range of advanced tuning techniques that rely on knowing the position of implanted electrodes relative to tonotopically mapped SG nerves. These techniques have the potential to improve hearing outcomes achieved with existing CIs. However, there has been no technology developed that allows accurate assessment of electrode position relative to stimulation targets in vivo. In this research, we will develop this technology and test a simple tuning scheme to assess its clinical utility. If successful, we will have developed an easy to use software package that has the potential to increase the effectiveness of existing cochlear implant technology and improve the quality of hearing restoration for implantees. To identify electrode position, existing techniques for identifying Cochlear Contour Advance electrode arrays in post-operative CT will be expanded for application to other types of electrodes. To identify stimulation targets (the SG), advanced active shape modeling techniques will be explored. These are powerful image processing methods that identify structures in images while constraining the shape of the results to be consistent with typical anatomical variations. The SG region will be tonotopically mapped using known heuristic equations that describe this frequency relationship. A simple tuning scheme will then be tested on implantees to evaluate the utility of this approach. Tuning parameters will include deciding which electrodes are on and off, determining the relative power of each electrode, and determining an electrode frequency allocation table. Positive results will demonstrate that these techniques will lead to more effective implant tuning and better hearing restoration. PUBLIC HEALTH RELEVANCE: The proposed project involves exploring algorithmic methods that will lead to performance optimizations of cochlear implant electrode arrays. Thus, the results of this project can potentially improve the quality of hearing restoration for cochlear implant users and improve efficiency of the tuning process.