Related Experiment Video
Updated: Jun 6, 2026
Technical Approach for Infrared Tracking for Soft Tissue Navigation with a Holographic Head-Mounted Display and Preclinical Validation
Published on: September 2, 2025
Dynamic uncertainty-level assessment framework for real-time needle tracking in CT-guided surgical environments.
Max Steiger1,2, Mohammad Rezapourian3,4, Marko Rak3
1Otto von Guericke University Magdeburg, Faculty of Computer Science, Chair of Virtual and Augmented Reality, Universitaetsplatz 2, 39106, Magdeburg, Germany. max.steiger@ovgu.de.
This study introduces a framework for real-time uncertainty assessment in computed tomography (CT)-guided interventions, improving needle tracking reliability. The system quantifies prediction trustworthiness, enhancing procedural safety and clinical outcomes.
Area of Science:
- Medical Imaging
- Interventional Radiology
- Computer Vision
Background:
- Accurate needle tracking is crucial for computed tomography (CT)-guided interventions.
- Current tracking systems lack real-time reliability quantification, impacting procedural safety.
- Clinicians need trustworthy guidance for successful interventions.
Purpose of the Study:
- To develop a Dynamic Uncertainty Level Assessment Framework for real-time tracking reliability estimation.
- To provide a quantitative, intuitive link between predicted uncertainty and spatial tracking error.
- To enhance trust and decision-making in CT-guided procedures.
Main Methods:
- Proposed a framework linking uncertainty score to spatial tracking error (0-100% scale).
- Implemented three approaches: weighted metrics, lightweight CNN, and hybrid CNN.
- Utilized multi-view image data for uncertainty prediction.
Main Results:
- Demonstrated strong positive correlation between uncertainty and error (Pearson r > 0.82).
- Achieved stable tracking error estimation (< 0.6 mm error difference).
- Real-time performance achieved at 5 ms per frame.
Conclusions:
- The framework enables intuitive uncertainty-to-error mapping for informed decision-making.
- Enhances operator trust in real-time guidance data.
- Provides a basis for integrating uncertainty-aware systems into CT-guided interventions.
