統計的特徴量を用いたアルツハイマー型痴呆診断支援システムに関する基礎研究

概要

論文の詳細を見る
In this study, in order to find out the possibility of Alzheimer-type dementia, we studied Alzheimer-type dementia with elderly volunteers using the run length matrix, on head MR images of 38 Alzheimer-type dementia patients (9 male, 29 female, 78.0 + 6.4 years) and 31 elderly volunteers (11 male, 20 female, 72.9 + 8.2 years) . The result of compared, parameter of gray level nonuniformity and run length nonuniformity from run length matrix and angular second moment, variance, inverse difference moment, sum average, sum variance, sum entropy, entropy, different variance, different entropy and information measure of correlation from co-occurrence were statistically significant. Furthermore, discriminant analysis based on 12 parameters resulted the sensitivity of 97.4%, the specificity of 96.8%, and the correct classification of 97.1%. From the result of this study, it has provided the possibility of MR image-based diagnosis of Alzheimer-type dementia with texture analysis.
長岡技術科学大学の論文
2003-02-13