Airway Navigational System

Period of Performance: 09/10/2007 - 03/09/2009


Phase 1 SBIR

Recipient Firm

Vida Diagnostics, Inc.
Principal Investigator


DESCRIPTION (provided by applicant): The detection of small nodules is possible given the emergence of high-resolution MDCT scanning, but often accurate diagnosis is delayed because a biopsy is not easily obtained. Thus, it is not possible to make the cancer diagnosis until rapid nodule growth is observed in subsequent follow up scans. This problem of finding nodules but delaying diagnosis and therapy is known as the lung cancer paradox. To assist in early lung-cancer detection, we propose a catheter-based system that can be accurately positioned inside the lung for traditional biopsy or optical sampling. In phase I, we will design a combined hardware and software system for planning a diagnostic procedure and guiding a catheter toward the precise region of interest using a magnetically-tracked sensor and a 3D model of the pulmonary anatomy derived from CT imaging. Also in phase I, we will begin to implement optical tracking of the endoscope tip as the foundation for our phase II hybrid system which will combine both magnetic and optical tracking into a single navigational tool. In this Phase I feasibility project, we propose four specific aims: Specific Aim 1: Refine the algorithms needed for the magnetic tracking of a bronchoscope tip during an endoscopic procedure. In particular, the phase one work includes refining the algorithms for: A. Visualization: Our navigational software's current graphical software displays a 3D model of a patient's airway created from CT-image data. In phase I, we will develop algorithms to include the patient's vascular tree in the visualization software. B. Registration: We will refine our current fiduciary point based registration to an easier to use touchless calibration process. Specific Aim 2: We will develop algorithms that limit error in reporting the bronchoscope's position due to patient respiratory motion. Specific Aim 3: We will develop algorithms to register endoscopic video to the patient's virtual 3D airway model, providing the foundation for our phase II hybrid system which will integrate magnetic and optical tracking. Specific Aim 4: Validate the system's ability to locate an airway target with less than a 5.0 mm mean error in position in a breathing lung, our phase one goal for accuracy. We will also measure how well the optical matching will improve the system's tracking accuracy. We will evaluate positioning accuracy of the magnetic tracking in an animal model as well as the positioning accuracy of our initial optical tracking algorithms. In the phase II work, we will develop a hybrid magneto-optical tracking technique that uses the bronchoscopic video to provide real time additional positioning accuracy and increased robustness in the presence of patient motion. This system will provide a unique platform to facilitate bronchoscopically-delivered procedures such as biopsy and non-surgical interventions for lung disease. Advances in novel, non-surgical therapies for lung disease make it critical to accurately target regions of interest and guide catheters and other devices into position for therapy. This project will combine a bronchoscope, magnetic position sensor, and a high resolution model of the thoracic anatomy into a pulmonary intervention workstation. Physicians can use this system to identify and quantify disease and regions of interest, plan in- terventions and other procedures, and precisely navigate toward regions of interest to obtain biopsies and/or deliver therapy.