Feature Selection by Genetic Algorithm for MRI Segmentation

被引：0

作者：

Matsui, Kazuhiro ^{[1
]}

Suganami, Yusuke ^{[1
]}

Kosugi, Yukio ^{[1
]}

机构：

[1] Interdisc. Grad. Sch. Sci. and Eng., Tokyo Institute of Technology, Yokohama, 226-8502, Japan

来源：

Systems and Computers in Japan | 1999年 / 30卷 / 07期

关键词：

D O I：

暂无

中图分类号：

学科分类号：

摘要：

To improve the performance of tissue-classification neural networks, it is important to select the optimal combination of feature indices, from among many candidates, to be used for the classification. In this paper, we propose a new method to select effective features using a genetic algorithm (GA). In our GA method, we use a new criterion, vector-quantized conditional class entropy (VQCCE), to evaluate the combination of feature indices rapidly without testing the actual classifiers. We applied our method for problems of brain MRI segmentation to classify gray matter/white matter regions. © 1999 Scripta Technica.

引用

页码：69 / 77