A support vector machine-based method to identify mild cognitive impairment with multi-level characteristics of magnetic resonance imaging
详细信息 查看全文
- 作者：Zhuqing Longa ; &dagger ; ; Bin Jinga ; &dagger ; ; Huagang Yana ; Jianxin Donga ; Han Liua ; Xiao Moa ; Ying Hanb ; c ; 13621011941@163.com" class="auth_mail" title="E-mail the corresponding author ; Haiyun Lia ; haiyunli@ccmu.edu.cn" class="auth_mail" title="E-mail the corresponding author
- 关键词：ALFF ; amplitude of low-frequency fluctuation ; AMYG ; amygdala ; AUC ; area under the curve ; BOLD ; blood oxygen level dependence ; CAU ; caudate ; CDR ; Clinical Dementia Rating scale ; DTI ; diffusion tensor imaging ; GMD ; gray matter density ; HE ; Hurst exponent ; HIP ; hippocampus ; IFG ; inferior frontal gyrus ; IPL ; inferior parietal lobule ; LING ; lingual gyrus ; LN ; lentiform nucleus ; LOOCV ; leave-one-out cross-validation ; MCI ; mild cognitive impairment ; MFG ; middle frontal gyrus ; MRI ; magnetic resonance im
- 刊名：Neuroscience
- 出版年：2016
- 出版时间：7 September 2016
- 年：2016
- 卷：331
- 期：Complete
- 页码：169-176
- 全文大小：1045 K
文摘
- •
Proposed a valid SVM-based method to identify MCI using HE, ALFF, ReHo and GMD.
- •
96.67% accuracy was achieved for identifying MCI from NCs.
- •
Single feature parameter could just obtain a maximum accuracy of 90.32%.
- •
Combined features could significantly improve the classification performance.
- •
The abnormal brain regions in MCI mainly involve several default mode regions.