TY - GEN
T1 - Exploring intrinsic functional differences of gyri, sulci and 2-hinge, 3-hinge joints on cerebral cortex
AU - Ge, Fangfei
AU - Zhang, Shu
AU - Huang, Heng
AU - Jiang, Xi
AU - Dong, Qinglin
AU - Guo, Lei
AU - Wang, Xianqiao
AU - Liu, Tianming
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - The human cerebral cortex has been commonly known as a highly-folded region which consists of convex gyri and concave sulci. Many previous studies have already revealed the fundamental differences of these convex and concave areas by analyzing structural and functional connectivity patterns. However, to our best knowledge, rare work has been done to explore their intrinsic functional differences from the perspective of neural activity, especially for 3-hinge gyral folding joints. Inspired by current evidences, in this paper, experiments based on classification models learned by convolutional neural network (CNN) are designed and performed on resting state functional magnetic resonance imaging (rsfMRI) data of both healthy controls and autism patients from the publicly available ABIDE II database. In our work, gyral and sulcal, 2-hinge and 3-hinge joint rsfMRI signals are modeled and predicted using CNNs with an average testing classification accuracy of 94.24% for controls, 95.24% for patients and 87.53% for controls, 87.72% for patients at individual level separately, which confirms different functional roles of neural activities under resting state in gyri and sulci, as well as 2-hinge and 3-hinge gyral folding joints in healthy subjects and autism groups. Besides, further analyses on learned characteristic features to differentiate gyral/sulcal, 2-hinge/3-hinge joint rsfMRI signals are also designed and performed to interpret our findings.
AB - The human cerebral cortex has been commonly known as a highly-folded region which consists of convex gyri and concave sulci. Many previous studies have already revealed the fundamental differences of these convex and concave areas by analyzing structural and functional connectivity patterns. However, to our best knowledge, rare work has been done to explore their intrinsic functional differences from the perspective of neural activity, especially for 3-hinge gyral folding joints. Inspired by current evidences, in this paper, experiments based on classification models learned by convolutional neural network (CNN) are designed and performed on resting state functional magnetic resonance imaging (rsfMRI) data of both healthy controls and autism patients from the publicly available ABIDE II database. In our work, gyral and sulcal, 2-hinge and 3-hinge joint rsfMRI signals are modeled and predicted using CNNs with an average testing classification accuracy of 94.24% for controls, 95.24% for patients and 87.53% for controls, 87.72% for patients at individual level separately, which confirms different functional roles of neural activities under resting state in gyri and sulci, as well as 2-hinge and 3-hinge gyral folding joints in healthy subjects and autism groups. Besides, further analyses on learned characteristic features to differentiate gyral/sulcal, 2-hinge/3-hinge joint rsfMRI signals are also designed and performed to interpret our findings.
KW - 2-hinge
KW - 3-hinge
KW - Autism
KW - Convolutional neural network
KW - Gyri
KW - Sulci
UR - http://www.scopus.com/inward/record.url?scp=85073888392&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2019.8759395
DO - 10.1109/ISBI.2019.8759395
M3 - 会议稿件
AN - SCOPUS:85073888392
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 1585
EP - 1589
BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
T2 - 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
Y2 - 8 April 2019 through 11 April 2019
ER -