期刊
BRAIN RESEARCH
卷 1402, 期 -, 页码 46-53出版社
ELSEVIER SCIENCE BV
DOI: 10.1016/j.brainres.2011.05.054
关键词
Heroin-dependent individuals; Pattern classification; Multi-voxel pattern analysis (MVPA); Non-negative matrix factorization (NMF); Support vector machine (SVM)
资金
- China Scholarship Council [2010696503]
- National Natural Science Foundation of China [60901064, 30970774, 30873462, 30870685, 81000641, 81000640, 81071217, 31028010, 81071137]
- Project for the National Key Basic Research and Development Program (973) [2011CB707700, 2011CB707702]
- CAS
- Fundamental Research Funds for the Central Universities
- NARSAD
Previous functional imaging studies on heroin addicts have focused on abnormal brain functions based on specific tasks, while few fMRI studies concentrated on the resting-state abnormalities of heroin-dependent individuals. In the current study, we applied the pattern classification technique, which employs the feature extraction method of non-negative matrix factorization (NMF) and a support vector machine (SVM) classifier. Its main purpose was to characterize the discrepancy in activation patterns between heroin-dependent individuals and healthy subjects during the resting state. The results displayed a high accuracy in the activation pattern differences of the two groups, which included the orbitofrontal cortex pm, cingulate gyrus, frontal and para-limbic regions such as the anterior cingulate cortex (ACC), hippocampal/parahippocampal region, amygdala, caudate, putamen, as well as the posterior insula and thalamus. These findings indicate that significant biomarkers exist among the network of circuits that are involved in drug abuse. The implications from our study may help explain the behavioral and neuropsychological deficits in heroin-dependent individuals and shed light on the mechanisms underlying heroin addiction. (C) 2011 Elsevier B.V. All rights reserved.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据