Abstract
Target delineation in Head and Neck cancer is a critical step in intensity modulated radiation therapy. Current methods to identify targets for delineation rely on volumetric imaging combined with other clinical findings including physical exam and endoscopy. Endoscopy and physical exam often reveal areas of visible or palpable disease which are not clearly demonstrated on volumetric imaging. However, relating endoscopy images to the volumetric imaging is subject to potential error relying on clinician recall of anatomy visible on both CT and endoscopy. We have developed technology that registers endoscopic images to CT images by tracking and registering the position and orientation of the endoscope relative to the CT image set using electromagnetic sensors embedded in the endoscope. After 2D to 3D image registration, users can contour regions‐of‐interest visible in the 2D endoscopic view. A mesh is created on the interior of the ROI and projected onto the 3D image data, registering the ROI with the volumetric image. This 3D ROI can be exported in a format that can be read by standard clinical treatment planning software. The technology was tested on an anatomical head phantom based on a subject image. Following CT imaging, endoscopy was performed on the head phantom. ROI's were contoured in the 2D endoscopic image, with points targeting known landmarks in the phantom. The ROI's were imported into Pinnacle. The accuracy of the ROI position was measured by comparison with the CT‐visible landmarks used to guide the initial 2D contouring, demonstrating a registration accuracy of ±1mm.
Original language | English (US) |
---|---|
Pages (from-to) | 3909 |
Number of pages | 1 |
Journal | Medical physics |
Volume | 37 |
Issue number | 7 |
DOIs | |
State | Published - Jul 2010 |
Externally published | Yes |
ASJC Scopus subject areas
- Biophysics
- Radiology Nuclear Medicine and imaging