Compressed Sensing based Tomosynthesis Image Reconstruction

Digital Tomosynthesis (DTS) is an innovative imaging modality that provides 3D reconstructed images of inner organs, using conventional digital x-ray equipment. The image slices are reconstructed from a series of projection images acquired as the x-ray tube moves around the patient in a limited angle path. DTS improves the sensitivity of lesion detection, due to reduced overlap of tissues. However, since the conventional reconstruction methods need many projections to achieve diagnostically high quality images, patients are exposed to higher amounts of radiation dose, which in turn increases the chance of cancer induction. This project aims to develop a new reconstruction algorithm in order to obtain high quality slice images using a relatively new mathematical theorem called Compressed Sensing (CS). CS enable high quality images to be reconstructed from few projections, leading to diagnostically better quality images derived from low dose scan protocols in a shorter time.

Faculty Supervisor:

Narinder Paul

Student:

Sayed Amroabadi

Partner:

Carestream Health Inc.

Discipline:

Engineering - biomedical

Sector:

Medical devices

University:

University of Toronto