Work

Locally Linear Embedding-Based Motion Estimation From Truncated Projections for Computed Tomography

Year: 2017

Type: article

Abstract: In the computed tomography (CT) field, image reconstructions from truncated projections acquired by only illuminating the region of interest are an effective method to reduce the radiation dose. Theoretically, it has been proven that the exact interior reconstruction is feasible with some prior knowledge. However, the traditional data-consistency-based motion correction methods cannot be applied to truncated data. In this paper, we propose a locally linear embedding (LLE)-based motion correction method for locally truncated projections. Compared with the fast rotation of the X-ray source, the object motion is relatively slow and can be approximated by a smooth polynomial function. Based on this knowledge, a constraint term is added to optimize the estimated motion parameters. Extensive numerical simulations are performed. Our results demonstrate the feasibility and satisfactory performance of the proposed method. As far as the authors know, this algorithm is the first of its kind for motion parameter estimation only from truncated projections in the CT field.

Source: IEEE Access

DOI: 10.1109/access.2017.2715011

Authors: Miaoshi Wang, Shuxu Guo, Hengyong Yu

Institutions: University of Massachusetts Lowell, Jilin University

Language: English

Topic: Medical Imaging Techniques and Applications

Subfield: Radiology, Nuclear Medicine and Imaging

Field: Medicine

Domain: Health Sciences

Open Access status: gold

Estimated APC paid: $2,075

Funder: Division of Chemical, Bioengineering, Environmental, and Transport Systems