TY - GEN
T1 - Multimodal neuroimaging patterns associated with social responsiveness impairment in autism
T2 - 16th IEEE International Symposium on Biomedical Imaging, ISBI 2019
AU - Li, Tiantian
AU - Fu, Zening
AU - Liu, Xia
AU - Qi, Shile
AU - Calhoun, Vince D.
AU - Sui, Jing
N1 - Funding Information:
This work was supported by China Natural Science Foundation (No. 61773380), the Strategic Priority Research Program of Chinese Academy of Science, Grant No. XDB03040100, Brain Science Program of Beijing City (No. Z181100001518005).
Publisher Copyright:
© 2019 IEEE.
PY - 2019/4
Y1 - 2019/4
N2 - Multimodal fusion is an effective approach to discover covarying patterns of multiple imaging types impaired in brain diseases, such as Autism spectrum disorders (ASD). Compelling evidence has confirmed that social impairment is a primary and common characteristic of ASD. Here, based on a supervised learning strategy and independent component analysis, we used Social Responsiveness Scale (SRS) of participants as reference, to guide the 3-way MRI data fusion, aiming to identify the multimodal brain regions associated with social impairment in ASD. Results show high consistence on the brain patterns identified in two independent cohorts, suggesting that lower fALFF and ReHo values in the default mode network (DMN) and occipital cortex, as well as reduced cortical thickness in inferior frontal and superior temporal cortex were associated with social impairment in autism. Moreover, these multimodal brain regions were also detected in a repeatable manner, validating the robustness of the fusion with reference approach, demonstrating the ability of multimodal fusion to identify potential imaging markers of interest for mental disorders.
AB - Multimodal fusion is an effective approach to discover covarying patterns of multiple imaging types impaired in brain diseases, such as Autism spectrum disorders (ASD). Compelling evidence has confirmed that social impairment is a primary and common characteristic of ASD. Here, based on a supervised learning strategy and independent component analysis, we used Social Responsiveness Scale (SRS) of participants as reference, to guide the 3-way MRI data fusion, aiming to identify the multimodal brain regions associated with social impairment in ASD. Results show high consistence on the brain patterns identified in two independent cohorts, suggesting that lower fALFF and ReHo values in the default mode network (DMN) and occipital cortex, as well as reduced cortical thickness in inferior frontal and superior temporal cortex were associated with social impairment in autism. Moreover, these multimodal brain regions were also detected in a repeatable manner, validating the robustness of the fusion with reference approach, demonstrating the ability of multimodal fusion to identify potential imaging markers of interest for mental disorders.
KW - Autism
KW - CCA #
KW - FMRI
KW - ICA
KW - Multimodal fusion>
KW - SMRI
UR - http://www.scopus.com/inward/record.url?scp=85073909074&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073909074&partnerID=8YFLogxK
U2 - 10.1109/ISBI.2019.8759460
DO - 10.1109/ISBI.2019.8759460
M3 - Conference contribution
AN - SCOPUS:85073909074
T3 - Proceedings - International Symposium on Biomedical Imaging
SP - 409
EP - 413
BT - ISBI 2019 - 2019 IEEE International Symposium on Biomedical Imaging
PB - IEEE Computer Society
Y2 - 8 April 2019 through 11 April 2019
ER -