The accuracy of a FEM model in predicting phantom deformation

Finished: 2021-03-31

MSc assignment

The MRI and Ultrasound Robotic Assisted Biopsy (MURAB) project aims to improve breast screening, more specifically the diagnosis of lesions that are only visible on MRI and not with ultrasound or mammography. After the initial MRI scan, a robot arm equipped with stereo camera, ultrasound and a moveable needle holder scans the breast after which MRI-US fusion takes place and the intervention is planned. This intervention involves positioning the ultrasound probe and inserting a biopsy needle to acquire a sample from the lesion, with real-time ultrasound feedback in which the MRI-visible lesion is projected.

Due to probe-tissue interactions the breast gets deformed, displacing the lesion and requiring the system to compensate for that to ensure good needle insertion accuracy. The amount of deformations can be simulated with FEM software. The central topic of this assignment is to validate these simulated deformations. How exactly you do this is up to you; one example is to track a finite number of markers on the skin in real-time using computer vision. Based on the results the estimation of the lesion displacement can be improved and the biopsy made more accurate.