当前位置: X-MOL 学术Neuroinformatics › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Functional Parcellation of Individual Cerebral Cortex Based on Functional MRI.
Neuroinformatics ( IF 3 ) Pub Date : 2019-12-04 , DOI: 10.1007/s12021-019-09445-8
Jiajia Zhao 1 , Chao Tang 1 , Jingxin Nie 1
Affiliation  

The human brain atlas assists us to enhance our scientific understanding of brain structure and function. The typical anatomical atlases are mainly based on brain morphometry which cannot ensure the consistency of structure and function, and are also hard to cover individual functional differences especially in cerebral cortex. Thus, in recent years, functional atlases for individuals have captured great attention, since they are essential not only for identifying the unique functional organization of individual brains, but also to explore individual variations in behaviors. In this study, a novel approach was proposed to accurately parcellate the whole cerebral cortex at the individual level using resting-state functional magnetic resonance image (rs-fMRI). To examine the functional homogeneity in parcellation, a new evaluation criterion, similarity of cluster (SC) coefficient, was proposed. The parcellation results demonstrated the high consistency between two resting-state sessions (Dice >0.72). The most consistent parcellation appeared in the frontal cortex and the least consistent parcellation appeared in the occipital cortex. The functional homogeneity of subregions was high in frontal cortex and insula whereas low in precentral gyrus. According to SC value, the optimal clustering number was about 1600 per hemisphere. Identification accuracy was 100% between two rs-fMRI sessions, and it was also above 0.97 for rest-task and task-task sessions.

中文翻译:

基于功能MRI的单个大脑皮层的功能分割。

人脑图谱有助于我们增强对脑结构和功能的科学理解。典型的解剖图谱主要基于大脑形态学,无法确保结构和功能的一致性,并且也难以覆盖各个功能的差异,尤其是在大脑皮层中。因此,近年来,针对个人的功能图谱引起了极大的关注,因为它们不仅对于识别个人大脑的独特功能组织至关重要,而且对于探索行为的个体差异都是必不可少的。在这项研究中,提出了一种新颖的方法,可以使用静止状态功能磁共振图像(rs-fMRI)在个体水平上准确分割整个大脑皮层。为了检查拼合中的功能同质性,新的评估标准,提出了集群相似度系数。分割结果证明了两个静止状态时段之间的高度一致性(Dice> 0.72)。最一致的切碎出现在额叶皮层,最不一致的切碎出现在枕叶皮层。次区域的功能同质性在额叶皮层和岛中较高,而在中央前回中较低。根据SC值,最佳聚类数约为每个半球1600。两次rs-fMRI会话之间的识别准确度为100%,其余任务和任务-任务会话的识别准确度也高于0.97。最一致的切碎出现在额叶皮层,最不一致的切碎出现在枕叶皮层。次区域的功能同质性在额叶皮层和岛中较高,而在中央前回中较低。根据SC值,最佳聚类数约为每个半球1600。两次rs-fMRI会话之间的识别准确度为100%,其余任务和任务-任务会话的识别准确度也高于0.97。最一致的切碎出现在额叶皮层,最不一致的切碎出现在枕叶皮层。次区域的功能同质性在额叶皮层和岛中较高,而在中央前回中较低。根据SC值,最佳聚类数约为每个半球1600。两次rs-fMRI会话之间的识别准确度为100%,其余任务和任务-任务会话的识别准确度也高于0.97。
更新日期:2019-12-04
down
wechat
bug