Analysis of Patient Customized Electrode Array Insertion Techniques in Cochlear Implant Surgery
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
In this dissertation, novel methods were developed that will facilitate the optimal placement of the electrode array during the cochlear implant (CI) procedure and thus improve the surgical outcome. CIs are considered to be an effective treatment for severe to profound hearing loss. During this procedure an electrode array is inserted into the cochlear region that stimulates the auditory nerve to create hearing sensation. CI procedures face several limitations, including the lack of visualization for the surgeon, which complicates accurate electrode placement. Variations in cochlear anatomy among individuals further challenge the procedure, leading to suboptimal array positioning. Major complications like fold-overs and scalar translocation can result in poor auditory outcomes, and over insertion into the cochlea's healthy region with residual hearing can cause damage to the healthy region and trauma. To address these limitations, we developed several image-guided pre-, intra- and post-operative approaches which are presented in this dissertation. To provide surgical guidance preoperatively, we have developed a framework for preoperative surgical guidance plan generation by analyzing the preoperative CT images. The guidance plan provides the necessary information, i.e., insertion depth, insertion trajectory and angle, that can help achieve optimal positioning of the electrode array. To detect electrode array fold-overs, one of the major complications during CI surgery, a weakly supervised deep learning-based model was developed. The model, trained with synthetic CT images, can detect fold-overs intraoperatively. In this dissertation, the sensitivity analysis of intra-cochlear anatomy segmentation methods to varying image acquisition parameters was also presented that can be used for both preoperative surgical planning and postoperative electrode programming. The methods presented in this dissertation can be utilized to improve the overall outcomes of the CI surgery and reduce the risk of complications and trauma.