Skip to main content

Medical-image Analysis and Statistical Interpretation Lab

SIMPLE Is a Good Idea (and Better with Context Learning)

Posted by on Tuesday, September 30, 2014 in Abdomen Imaging, Image Segmentation.

Zhoubing Xu, Andrew J. Asman, Peter L. Shanahan, Richard G. Abramson, Bennett A. Landman. “SIMPLE Is a Good Idea (and Better with Context Learning)”, In International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), Boston, MA, September 2014.

Full text: PubMed

Abstract

Selective and iterative method for performance level estimation (SIMPLE) is a multi-atlas segmentation technique that integrates atlas selection and label fusion that has proven effective for radiotherapy planning. Herein, we revisit atlas selection and fusion techniques in the context of segmenting the spleen in metastatic liver cancer patients with possible splenomegaly using clinically acquired computed tomography (CT). We re-derive the SIMPLE algorithm in the context of the statistical literature, and show that the atlas selection criteria rest on newly presented principled likelihood models. We show that SIMPLE performance can be improved by accounting for exogenous information through Bayesian priors (so called context learning). These innovations are integrated with the joint label fusion approach to reduce the impact of correlated errors among selected atlases. In a study of 65 subjects, the spleen was segmented with median Dice similarity coefficient of 0.93 and a mean surface distance error of 2.2 mm.

Flowchart of the proposed method. Given registered atlases with variable qualities, atlas selection and statistical fusion are considered as two necessary steps to obtain a reasonable fusion estimate of the target segmentation. The SIMPLE algorithm implicitly combines these two steps to fusion selected atlases; however, more information can be incorporated to improve the atlas segmentation, and a more advanced fusion technique can be used after the atlases are selected. We propose to (1) extract a probabilistic prior of the target segmentation by context learning to regularize the atlas selection in SIMPLE and (2) use Joint Label Fusion to obtain the final segmentation while characterizing of the correlated errors from among atlases.
Flowchart of the proposed method. Given registered atlases with variable qualities, atlas selection and statistical fusion are considered as two necessary steps to obtain a reasonable fusion estimate of the target segmentation. The SIMPLE algorithm implicitly combines these two steps to fusion selected atlases; however, more information can be incorporated to improve the atlas segmentation, and a more advanced fusion technique can be used after the atlases are selected. We propose to (1) extract a probabilistic prior of the target segmentation by context learning to regularize the atlas selection in SIMPLE and (2) use Joint Label Fusion to obtain the final segmentation while characterizing of the correlated errors from among atlases.

Tags: