Automatic CT-to-patient Registration During Navigated Bronchoscopy and EBUS

Bronchoscopic diagnosis and staging of lung cancer must be gentle and quick, and requires optimal sampling precision.Electromagnetic navigation systems are promising for intraoperative guiding based on maps made of preoperative CT images. Navigation accuracy depends largely on correct alignment of preoperative images to the patient´s position in space during endoscopy. In this study, the accuracy of a new fast and automatic image-to-patient registration method during the initial phase of bronchoscopy is assessed in patients referred for lung cancer investigation.

This is an accuracy study of a new automatic CT-to-patient registration technique for bronchoscopy and endobronchial ultrasound bronchoscopy (EBUS). An in-house research and development navigation platform displays the intraoperative position of the endoscope inside the airways, and shows its exact position in three-dimensional (3D) maps made from the patients' own preoperative CT images. The navigation system is used for acquiring position data from a bronchoscope or EBUS-scope (same sort of output data acquired from both endoscopes in this study) equipped with a position sensor for electromagnetic tracking. From intraoperative position data, the accuracy of the image-to-patient registration and the navigation system accuracy can be calculated. When targeting a certain lymph node or tumor for sampling, the pulmonologist can use electromagnetic navigation combined with preoperative CT images to navigate directly, precisely and accurately to the area of interest. Functional systems for navigated bronchoscopy and EBUS may therefore lead to considerable improvements in lung cancer diagnosis and mediastinal staging.