DALM, Deformable Attenuation-Labeled Mesh for Tomographic Reconstruction and Segmentation

Jakeoung Koo,Vedrana Anderson Dahl,Anders Bjorholm Dahl

doi:10.1109/tci.2021.3052034

Jakeoung Koo, Vedrana Anderson Dahl + Show 1 more

Open Access

https://doi.org/10.1109/tci.2021.3052034

Copy DOI

Abstract

Most X-ray tomographic reconstruction methods represent a solution as an image on a regular grid. Such representation may be inefficient for reconstructing homogeneous objects from noisy or incomplete projections. Here, we propose a mesh-based method for reconstruction and segmentation of homogeneous objects directly from sinogram data. The outcome of our proposed method consists of curves outlining the regions of constant attenuation, and this output is represented using a labeled irregular triangle mesh. We find the solution by deforming the mesh to minimize the residual given by the sinogram data. Our method supports multiple materials, and allows for topological changes during deformation. An integral part of our algorithm is an efficient forward projection of the labeled mesh onto the sinogram domain. We initialize our algorithm based on graph total variation, also here taking advantage of the mesh representation. Experimental results on simulated datasets show that our method gives a compact representation of the reconstruction and also accurate segmentation results for challenging data with e.g. large noise, a small number of angles or problems with limited angle. We also demonstrate the result on real fan-beam data. The proposed geometric solution shows a further step towards using alternative representations for tomographic reconstruction.

Highlights

I N X-RAY CT, the aim of the tomographic reconstruction is typically to compute an image of a scanned object with image voxels representing attenuation coefficients arranged on a regular grid [1]
We show the relative advantages over other reconstruction methods on various synthetic datasets
We have presented deformable attenuation-labeled mesh (DALM), a mesh-based method for 2D reconstruction and segmentation directly from sinogram data

Summary

Introduction

I N X-RAY CT, the aim of the tomographic reconstruction is typically to compute an image of a scanned object with image voxels representing attenuation coefficients arranged on a regular grid [1]. This representation is well-suited for a wide range of reconstruction problems. We propose a reconstruction and segmentation method using an interface-based representation in the form of a labeled deformable mesh and we investigate the properties of our method. Our work extends the method by Dahl et al [2] that employs a deformable closed curve to outline one object in the reconstruction.

Objectives

Methods

Results

Conclusion