Automated Delineation of Regional Target Volumes for Patients with Head and Neck Cancer Treated Conformally
Last Modified: October 7, 2002
Presenter: J. Barker
Presenter's Affiliation: Department of Radiation Oncology, University of Washington, Seattle, WA, USA
Type of Session: Scientific
- One of main challenges in designing radiation treament fields is accurately identifying anatomic structures on planning radiographs.
- This issue is becoming even more important when conformal techniques are utilized and doses are escalated.
- In the current study, investigators attempt to design a computer algorithm that can accurately and automatically contour cevical lymph nodes.
- A library of head and neck CT scans was amassed for the reference model.
- With the assistance of a radiologist, neck lymph node stations were identified and contoured on each axial slice of each scan.
- A sophisticated imaging algorithm was then developed to allow individual CT scans to be compared (or co-registered) with the reference data.
- Based on this data, the program then contours neck lymph nodes on the individual patient's CT scan.
- During the presentation images of this process were displayed.
- Noted, was that the patient's CT scan gets "transformed" to fit the computer model. In other words, individual scans get altered from there original diagnostic representation. This makes visual interpretation by the clinician more complex.
- The authors concluded that while the teqnique is promising, it is not yet ready for clinical use.
- They indicate that more refinements of the algorithm are planned.
- This presentation revealed a remarkable new technology which is being developed to assist clinicians in the design of radiation treatment fields.
- Potentially, this system could enhance treatment results by allowing for more accurate definition of at-risk lymph node regions.
Oncolink's ASTRO Coverage made possible by an unrestricted Educational Grant from Ortho Biotech.