A Patient-Specific Correspondence Model to Track Tumor Location in Thorax During Radiation Therapy

Abstract

<h3>Purpose/Objective(s)</h3> In radiation therapy, accurate localization of tumors is crucial to deliver the maximum dose to the tumor while sparing the organs at risk. However, respiration-induced motion creates challenges in the localization of the tumors in the thorax and abdomen. In this study, we present a patient-specific correspondence model to track tumors in the thorax during radiation therapy using surface displacement as the surrogate signal. <h3>Materials/Methods</h3> The proposed model is made prior to the treatment for each patient. Four-dimensional computed tomography (4D-CT) images of the patient are used as the knowledge of internal motion, and the displacement of two points on the patient's skin on the thoracic area is used as surrogate signals. The two types of data are acquired simultaneously. The 4D-CT images are sorted into ten respiratory phases by the amplitude-binning algorithm. A deformable image registration algorithm is applied to the 4D-CT images to extract the patient's internal motion data. The patient's average skin surface displacement at each respiratory phase is calculated from surrogate signals. Principal component analysis is used to fit the correspondence model. During radiation delivery, the model incorporates recorded surrogate signals as an input and delivers the 3D trajectory of the tumor or other anatomy of interest. We evaluated the accuracy of the proposed model on a respiratory phantom and five lung cancer patients. <h3>Results</h3> For the respiratory phantom, the location of the center of a target sphere inside the phantom during treatment was calculated in three directions: Left-Right (LR), Anterior-Posterior (AP) and, Superior-Inferior (SI). The error of localization was less than 1 mm in the LR and AP directions and less than 2 mm in the SI direction. For two patients, the location of the treatment isocenter and for the other three patients, the location of the apex of the diaphragm was calculated in three directions. The localization error in the LR and AP directions was less than 1 mm for all patients. For four of the patients, the localization error in the SI direction was less than 2 mm, and for one patient, it was less than 3 mm. <h3>Conclusion</h3> To reduce the unnecessary radiation exposure to the organs at risk, accurate localization of the tumors that have movement due to respiration is essential. In this study, we propose a patient-specific model to locate the tumor or any anatomy of interest in the thorax during radiation treatment. Our model is dose-free and accounts for hysteresis and irregular breathing.