From Preoperative CT to Postmastoidectomy Mesh Construction: Mastoidectomy Shape Prediction for Cochlear Implant Surgery

2026-01-07

arXiv.org

AI Summary

Preoperative CT imaging struggles to accurately predict the mastoidectomy cavity, limiting precise surgical planning for cochlear implantation. To address this challenge, this work proposes a novel framework that integrates self-supervised and weakly supervised learning to directly predict the postoperative resection shape from preoperative CT scans of intact mastoids—without requiring manual annotations. The method introduces an innovative 3D t-distribution loss function that significantly enhances the modeling of anatomical structures with ambiguous boundaries under weak supervision. Experimental results demonstrate that the proposed approach achieves a mean Dice score of 0.72 in predicting complex mastoidectomy regions, outperforming existing methods and laying a foundation for accurate postoperative 3D reconstruction.

Technology Category

• Machine Learning: Learning with Manifolds
• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Cochlear Implant (CI) surgery treats severe hearing loss by inserting an electrode array into the cochlea to stimulate the auditory nerve. An important step in this procedure is mastoidectomy, which removes part of the mastoid region of the temporal bone to provide surgical access. Accurate mastoidectomy shape prediction from preoperative imaging improves pre-surgical planning, reduces risks, and enhances surgical outcomes. Despite its importance, there are limited deep-learning-based studies regarding this topic due to the challenges of acquiring ground-truth labels. We address this gap by investigating self-supervised and weakly-supervised learning models to predict the mastoidectomy region without human annotations. We propose a hybrid self-supervised and weakly-supervised learning framework to predict the mastoidectomy region directly from preoperative CT scans, where the mastoid remains intact. Our hybrid method achieves a mean Dice score of 0.72 when predicting the complex and boundary-less mastoidectomy shape, surpassing state-of-the-art approaches and demonstrating strong performance. The method provides groundwork for constructing 3D postmastoidectomy surfaces directly from the corresponding preoperative CT scans. To our knowledge, this is the first work that integrating self-supervised and weakly-supervised learning for mastoidectomy shape prediction, offering a robust and efficient solution for CI surgical planning while leveraging 3D T-distribution loss in weakly-supervised medical imaging.

Problem

• mastoidectomy shape prediction
• cochlear implant surgery
• preoperative CT
• surgical planning
• medical image analysis

Innovation

• self-supervised learning
• weakly-supervised learning
• mastoidectomy shape prediction
• 3D T-distribution loss
• cochlear implant surgery