Monocular Marker-free Patient-to-Image Intraoperative Registration for Cochlear Implant Surgery

2025-05-23

AI Summary

Conventional intraoperative registration methods for cochlear implantation—relying on external optical tracking systems or surface fiducial markers—suffer from complex setup and poor real-time performance. To address this, we propose a zero-shot, monocular, markerless, and hardware-free 6D camera pose estimation framework. Our approach employs a lightweight neural network to directly map preoperative CT-derived 3D anatomical structures to intraoperative microscope 2D images, enabling real-time patient-to-image registration. Trained exclusively on synthetic microscopic surgical data, the model regresses full 6D pose parameters (rotation matrix and translation vector) and demonstrates strong cross-patient generalization. Evaluated on nine clinical cases, it achieves clinically viable accuracy, with >85% of angular errors ≤10°. This work presents the first monocular, intraoperative registration method for cochlear implantation that eliminates the need for optical trackers or fiducial markers—significantly enhancing the real-time capability, robustness, and clinical deployability of surgical navigation systems.

Technology Category

• Computer Vision: Motion & Tracking
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements

Abstract

This paper presents a novel method for monocular patient-to-image intraoperative registration, specifically designed to operate without any external hardware tracking equipment or fiducial point markers. Leveraging a synthetic microscopy surgical scene dataset with a wide range of transformations, our approach directly maps preoperative CT scans to 2D intraoperative surgical frames through a lightweight neural network for real-time cochlear implant surgery guidance via a zero-shot learning approach. Unlike traditional methods, our framework seamlessly integrates with monocular surgical microscopes, making it highly practical for clinical use without additional hardware dependencies and requirements. Our method estimates camera poses, which include a rotation matrix and a translation vector, by learning from the synthetic dataset, enabling accurate and efficient intraoperative registration. The proposed framework was evaluated on nine clinical cases using a patient-specific and cross-patient validation strategy. Our results suggest that our approach achieves clinically relevant accuracy in predicting 6D camera poses for registering 3D preoperative CT scans to 2D surgical scenes with an angular error within 10 degrees in most cases, while also addressing limitations of traditional methods, such as reliance on external tracking systems or fiducial markers.

Problem

• Monocular marker-free registration for cochlear implant surgery
• Mapping CT scans to 2D surgical frames via lightweight neural network
• Estimating camera poses without external tracking hardware

Innovation

• Monocular marker-free registration without external hardware
• Lightweight neural network for real-time CT to 2D mapping
• Zero-shot learning with synthetic surgical scene dataset