SPARSE: Seed Point Auto-Generation for Random Walks Segmentation Enhancement in medical inhomogeneous targets delineation of morphological MR and CT images.

Haibin Chen,Xuejun Gu,Xin Zhen,Laura Cervino,Yang Xiao,Hao Yan,Linghong Zhou

doi:10.1120/jacmp.v16i2.5324

Haibin Chen, Xuejun Gu + Show 5 more

Open Access

PDF Available

https://doi.org/10.1120/jacmp.v16i2.5324

Copy DOI

Export

Save

Cite

Abstract
Highlights/Summary
Full-Text PDF
Similar Papers

Abstract

Listen

In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user‐labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.PACS number: 87.57.nm

Highlights

Segmentation of medical images is a significant and challenging task for disease diagnosis[1,2] and treatment planning.[3,4,5] It can be roughly classified into three categories, namely, manual, semiautomatic, and automatic segmentation.[6]
It has been demonstrated that random walks (RW) is stable to changes of the seed points (SPs) only when the changes of location and quantity are below 10% and 50%, respectively.[9] to obtain satisfactory segmentation results, the quantity of initial SPs should be sufficient to be representative of almost all intensity levels in the target
We presented a RW-based segmentation algorithm, SPARSE, which incorporates a novel SPs autogeneration scheme for segmentation of inhomogeneous targets

Summary

Introduction

Segmentation of medical images is a significant and challenging task for disease diagnosis[1,2] and treatment planning.[3,4,5] It can be roughly classified into three categories, namely, manual, semiautomatic (interactive), and automatic segmentation.[6]. It has been demonstrated that RW is stable to changes of the SPs only when the changes of location and quantity are below 10% and 50%, respectively.[9] to obtain satisfactory segmentation results, the quantity of initial SPs should be sufficient to be representative of almost all intensity levels in the target. In practice, it is tedious and laborious to place sufficient SPs on the target, especially in a 3D scenario. In this sense, a stable segmentation result may not be guaranteed when insufficient SPs are used, especially for medical inhomogeneous targets, which are common in clinical studies

Methods

Results

Conclusion