Curative radiotherapy of localised cancer delivers discrete doses of radiation to diseased tissue over many days. Computer driven multileaf collimators now closely conform radiation dose fields to the shape of any tumour. This technology invites the reduction of beam margins in order to maximally spare healthy tissue, reducing sideeffects, but placing more emphasis on the precision of dose delivery. However, tumour motion and body surface variation demand more accurate patient set-ups.Software has been developed to assist in the task of aligning a patient on the treatment couch with his/her previously planned CT scan in true three-dimensions. A body surface height-map is constructed from a B-Spline skinned model of the planned CT scan outline set and compared with a height-map from an optical sensor. This type of co-registration does not depend entirely on skin surface markers which unavoidably move due to the dynamics of internal/external anatomical forces.
Conformal radiotherapy uses multi-leaf collimation to customise radiation dose fields to the three-dimensional shape of the target tumour. Its effectiveness is governed by the precision of dose delivery achieved using a few coplanar tattoo markers. During a course of treatment, the patient's body surface profile fluctuates causing the reference skin tattoo markers to move and hence, without the ability to verify target position, the precision of pretreatment set-ups will degrade, often significantly.Cross-modal patient set-up verification software has been developed which allows the user to examine images from various modalities. All imaging modalities available can be used to collectively define anatomical landmarks, outlined on a representative reference image in order to verify radiation field placement using anterior and lateral pre-treatment portal images. This software is presently being used within a national clinical trial for recording the field placement error observed daily at pre-treatment.
Recent advances in radical broad-beam conformal radiotherapy for disease at nonrigid sites suggest that the treatment room set-up of patients should be performed using the CT-plan digital body surface as a reference standard. In effect, the CT-plan surface is a "virtual shell" and should be used as such. T o do this requires 3D body surface information to be captured live in the treatment room and then manipulated for comparison with the C T virtual shell.T o address this situation, a near real-time prototype opto-electronic dynamic 3D surface sensor has been developed and equipped with a novel software utility for visualizing and matching the disparate surfaces arising from planning and treatment environments. Using the C T virtual shell and height-maps produced by the sensor, a two-phase 3D positioning strategy for patient set-up and monitoring is described. Phase one is an image-guided manual "approach". Phase two is automated "docking" using simulated annealing to determine optimum set-up corrections. The concepts and implementation are illustrated using archival patient data and an anthropomorphic phantom "patient" in a treatment room environment. Comp Aid Surg 5: 234-245 (2000). 02000 Wiley-Liss, hc.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.