Real-time simulation framework for lung tumor radiation therapy

Lung tumors move during breathing depending on patient's breathing conditions, thereby compromising the accurate deposition of radiation doses. It is thus important to calculate the delivered dose to various aspects of the moving tumor and the surrounding normal tissues due to change in tumor shape and location with breathing. In this abstract, we present a computer-based simulation framework that models a volumetric lung tumor, simulates the tumor motion during radiation therapy, and predicts the amount and location of radiation doses deposited in the moving lung tumor during the actual delivery of radiation. It will also provide insights on the variations in the effectiveness of the therapy for changes in the patient's breathing conditions. The simulation framework concists of a virtual cuboid of dimension 100X100X100 cubes created with each cube of the dimension 1X1X1 cubic mm. The 3D model (spherical, for validation purposes) of the tumor is introduced inside the cuboid. For every breathing step, the centroid of the lung tumor is translated in a randomly selected trajectory. Each 3D vertex inside the moving tumor is traced within the cuboid using binary level-set searching algorithm. To simulate radiation dose delivered, a radiation treatment plan of a small lung tumor was developed in a commercial planning system (BrainScan software, BrainLab). The dose for each radiation field was extracted as a 10cm cube to match the above described simulation cuboid. During the simulation of lung tumor motion, the dose on the target was summed to generate real-time dose to the target for each beam independently. The simulation results are validated by film dosimetry measurements using a physical lung phantom with a moving spherical tumor. Introduction The focus of the current work is for enhancing radiation therapy for lung tumors. Lung tumors move unpredictably depending on patient breathing patterns, thereby changing tumor location that subsequently compromises the accurate deposition of radiation doses. This study involved developing a real-time simulation method to calculate the delivered dose to various aspects of the moving tumor. Such real-time simulations of the actual location and shape of the tumor during the delivery of radiation would enable the use of high focused radiation fields that could result in decreased treatment related toxicitiy. The simulation framework takes into account the patient specific lung tumor motion extracted from Computed Tomography images and the radiation plan prescribed for the patient. The output of the simulation framework would predict the amount and location of …