Modeling of a Patient Positioning System for use in MRI Machines

Daniel Garcia, Marquette University


Since its commercial introduction in the early 1980s, Magnetic Resonance Imaging (MRI) has become an important medical diagnostic tool for radiologists. Researchers and manufacturers have refined the imaging hardware and expanded the intended uses for MRI devices over time. However, MRI manufacturers have not improved the design of the mechanisms and control schemes used to move the patient. Patient positioning systems are required to handle unknown weights up to 225 kilograms, accommodate friction disturbances, move long distances at high speed, move small distances in less than 1 second, and attain sub millimeter bidirectional precision. Little research is available on model based design of patient positioning mechanisms. A representative model of an MRI patient positioning device is proposed. The novelty of patient positioning system design in a strong magnetic field is examined in detail. Methods for creating an appropriate fidelity high level system friction, single degree of freedom Simulink model of the system are discussed. A physical facsimile of the system is constructed. The output of the Simulink model is compared to the physical system performance with respect to robustness against friction coefficient sensitivity, load mismatches and noise disturbances. Potential model simplifications and future work opportunities are highlighted.