0197
View Abstract / View Presentation
Data-driven clustering differentiates subtypes of major depressive disorder with distinct connectivity-symptom association
Yanlin Wang1, Shi Tang1, Xinyu Hu1, Yongbo Hu1, Weihong Kuang2, Zhiyun Jia1, Xiaoqi Huang1, and Qiyong Gong1
1Department of Radiology, West China Hospital, Sichuan University, Huaxi MR Research Center (HMRRC), Functional and molecular imaging Key Laboratory of Sichuan Province, Chengdu, China, 2Department of Psychiatry, Sichuan University West China Hospital, Chengdu, China
We defined a novel two specific symptoms dominated subtypes of depression that were underlined by distinct network patterns, which did not only improve our understanding of heterogeneous but also lead to the development of leveraging multimodal data types to define subtypes of depression.
Figure 1. Overview of data processing and analysis pipeline. Group-ICA was used to calculate and extract the average signal time series from validated brain parcellations. Following data reduction were performed by a rCCA. The subtypes of depression were defined by k-means clustering based on combination of neuroimaging and clinical data. Finally, group comparisons of ICNs and clinical symptoms between each subgroup and HC were implemented a one-way ANOVA analysis, respectively. PLSR analysis was employed to assess the association between these features in each subgroup.
Figure 4. Statistically significant FNC differences between MDD, MDD-subtypes and HCs. (A) A heatmap and a circular plot showing were significantly abnormal FNC in MDD relative to HCs. (B) A heatmap showing the significant results of ANOVA in FNC among MDD-subtypes and HCs. (C) Three circular plots showing significant FNC patterns between subtype1, subtype2 and HCs using post-hoc analyses. Statistical significance for all comparisons were thresholded at p < 0.05, FDR corrected for cluster level and p < 0.001, uncorrected for voxel level.