Model-based Catheter Shape Reconstruction for Interventional MRI

Introduction The use of real-time MRI for guidance of intravascular interventions is a rapidly developing research field given its superior soft-tissue contrast and the lack of hazardous radiation [1]. In order to visualize the entire length of such a catheter, a three dimensional (3D) image of a complete volume covering the catheter is necessary. Conventional MR imaging methods result in prohibitively long measurement times hampering 3D imaging in real-time. This may be addressed by applying spare signal sampling theorems [2] to 3D MR imaging of catheter shapes [3]. Despite progress in this direction, undersampling factors remain limited to reconstruct images at sufficient quality for performing 3D curve fit of the catheter shape. Recently, a reconstruction method was proposed to determine parameters of perfusion model directly from undersampled k-space data [4]. In the present work a model-based reconstruction technique is proposed to determine the parameters of a 3D catheter shape model from undersampling k-space data taking into account the limited degrees of freedom. For this, a parameterized model of a catheter is fitted to the acquired k-space data by minimizing the l2-norm. Using computer simulations and phantom data it is demonstrated that the catheter shape can be reconstructed from highly undersampled data indicating the potential of the method for 3D real-time imaging of catheter devices.