Development and comparison of new hybrid motion tracking for bronchoscopic navigation

This paper presents a new hybrid camera motion tracking method for bronchoscopic navigation combining SIFT, epipolar geometry analysis, Kalman filtering, and image registration. In a thorough evaluation, we compare it to state-of-the-art tracking methods. Our hybrid algorithm for predicting bronchoscope motion uses SIFT features and epipolar constraints to obtain an estimate for inter-frame pose displacements and Kalman filtering to find an estimate for the magnitude of the motion. We then execute bronchoscope tracking by performing image registration initialized by these estimates. This procedure registers the actual bronchoscopic video and the virtual camera images generated from 3D chest CT data taken prior to bronchoscopic examination for continuous bronchoscopic navigation. A comparative assessment of our new method and the state-of-the-art methods is performed on actual patient data and phantom data. Experimental results from both datasets demonstrate a significant performance boost of navigation using our new method. Our hybrid method is a promising means for bronchoscope tracking, and outperforms other methods based solely on Kalman filtering or image features and image registration.

The proposed hybrid motion prediction method comprises main three stages: feature-based motion prediction, Kalman filtering-based scale factor estimation, and intensity-based image registration. Inputs to the tracking method are bronchoscopic videos and CT datasets. Outputs of the tracking method are a series of the bronchoscope pose parameters (camera trajectories) or virtual bronchoscopic images corresponding to real bronchoscopic frames.Figure optionsDownload high-quality image (118 K)Download as PowerPoint slideResearch highlights
► Development of navigated bronchoscopy for diagnosis and treatment of lung cancer.
► Synchronization between 3D pre-built model and 2D bronchoscopic camera views.
► A new hybrid tracking method bases on SIFT, Kalman filter, and image registration.
► Our approach provides a more accurate and robustness bronchoscope tracking.