TY - JOUR
T1 - Modeling functional difference between gyri and sulci within intrinsic connectivity networks
AU - Wang, Qiyu
AU - Zhao, Shijie
AU - He, Zhibin
AU - Zhang, Shu
AU - Jiang, Xi
AU - Zhang, Tuo
AU - Liu, Tianming
AU - Liu, Cirong
AU - Han, Junwei
N1 - Publisher Copyright:
© 2022 The Author(s). Published by Oxford University Press. All rights reserved.
PY - 2023/2/15
Y1 - 2023/2/15
N2 - Recently, the functional roles of the human cortical folding patterns have attracted increasing interest in the neuroimaging community. However, most existing studies have focused on the gyro-sulcal functional relationship on a whole-brain scale but possibly overlooked the localized and subtle functional differences of brain networks. Actually, accumulating evidences suggest that functional brain networks are the basic unit to realize the brain function; thus, the functional relationships between gyri and sulci still need to be further explored within different functional brain networks. Inspired by these evidences, we proposed a novel intrinsic connectivity network (ICN)-guided pooling-trimmed convolutional neural network (I-ptFCN) to revisit the functional difference between gyri and sulci. By testing the proposed model on the task functional magnetic resonance imaging (fMRI) datasets of the Human Connectome Project, we found that the classification accuracy of gyral and sulcal fMRI signals varied significantly for different ICNs, indicating functional heterogeneity of cortical folding patterns in different brain networks. The heterogeneity may be contributed by sulci, as only sulcal signals show heterogeneous frequency features across different ICNs, whereas the frequency features of gyri are homogeneous. These results offer novel insights into the functional difference between gyri and sulci and enlighten the functional roles of cortical folding patterns.
AB - Recently, the functional roles of the human cortical folding patterns have attracted increasing interest in the neuroimaging community. However, most existing studies have focused on the gyro-sulcal functional relationship on a whole-brain scale but possibly overlooked the localized and subtle functional differences of brain networks. Actually, accumulating evidences suggest that functional brain networks are the basic unit to realize the brain function; thus, the functional relationships between gyri and sulci still need to be further explored within different functional brain networks. Inspired by these evidences, we proposed a novel intrinsic connectivity network (ICN)-guided pooling-trimmed convolutional neural network (I-ptFCN) to revisit the functional difference between gyri and sulci. By testing the proposed model on the task functional magnetic resonance imaging (fMRI) datasets of the Human Connectome Project, we found that the classification accuracy of gyral and sulcal fMRI signals varied significantly for different ICNs, indicating functional heterogeneity of cortical folding patterns in different brain networks. The heterogeneity may be contributed by sulci, as only sulcal signals show heterogeneous frequency features across different ICNs, whereas the frequency features of gyri are homogeneous. These results offer novel insights into the functional difference between gyri and sulci and enlighten the functional roles of cortical folding patterns.
KW - cerebral cortex
KW - cortical folding complexity
KW - fully convolutional neural network
KW - functional brain networks
UR - http://www.scopus.com/inward/record.url?scp=85133914567&partnerID=8YFLogxK
U2 - 10.1093/cercor/bhac111
DO - 10.1093/cercor/bhac111
M3 - 文章
C2 - 35332916
AN - SCOPUS:85133914567
SN - 1047-3211
VL - 33
SP - 933
EP - 947
JO - Cerebral Cortex
JF - Cerebral Cortex
IS - 4
ER -