High accuracy patient positioning system: geometric and elastic error calibration of a flexible model

Abstract

Radiotherapy and proton therapy, successfully used in cancer cure, are medical treatments that require very high positioning accuracy. The main goal of the POROS project is to demonstrate that patient positioning systems (PPS) based on industrial robots are reliable solutions for these treatments. However, due to its high number of degrees of freedom (6 DOF), size, and especially important payload weight, such an industrial robot, rigid by definition, holds elastic and geometric deformations. A method to calibrate this type of system in order to take into account these deformation errors, for different payloads, is presented in this article. As the patient weight varies and can reach high values, the rigid model is completed by a flexible one taking into account these deformations. Then, this model with a physical meaning of its parameters is calibrated, in order to reduce the positioning errors. The results of the rigid and the flexible model are compared, indicating that a flexible model is more adapted for the high accuracy requirements of a PPS. Validation results confirm the effectiveness of this model.