Functional and Structural Hierarchy in Individuals at Ultra High Risk for Developing Psychosis
Poster No:
505
Submission Type:
Abstract Submission
Authors:
Thuan Tinh Nguyen1,2,3, Siwei Liu1,2, Chen Hao Wang1, Michael Chee1,2, Jimmy Lee4, Juan Helen Zhou1,2,5,6
Institutions:
1Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore, 2Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Singapore, 3Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of Singapore, Singapore, Singapore, 4Department of Psychosis, Institute of Mental Health, Singapore, Singapore, 5Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of S, Singapore, Singapore, 6Department of Electrical and Computer Engineering, National University of Singapore, Singapore, Singapore
First Author:
Thuan Tinh Nguyen
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore|Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of Singapore
Singapore, Singapore|Singapore, Singapore|Singapore, Singapore
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore|Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of Singapore
Singapore, Singapore|Singapore, Singapore|Singapore, Singapore
Co-Author(s):
Siwei Liu
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore|Singapore, Singapore
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore|Singapore, Singapore
Chen Hao Wang
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore
Michael Chee
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore|Singapore, Singapore
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore
Singapore, Singapore|Singapore, Singapore
Juan Helen Zhou
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore|Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of S|Department of Electrical and Computer Engineering, National University of Singapore
Singapore, Singapore|Singapore, Singapore|Singapore, Singapore|Singapore, Singapore
Centre for Sleep and Cognition, Yong Loo Lin School of Medicine, National University of Singapore|Centre for Translational Magnetic Resonance Research, Yong Loo Lin School of Medicine, National University of Singapore|Integrative Sciences and Engineering Programme (ISEP), NUS Graduate School, National University of S|Department of Electrical and Computer Engineering, National University of Singapore
Singapore, Singapore|Singapore, Singapore|Singapore, Singapore|Singapore, Singapore
Introduction:
Research on individuals at ultra high risk for psychosis (UHR) demonstrates grey matter reduction and brain functional dysconnectivity, suggesting potential for predicting psychosis transition [1, 2]. Gradient-based studies showed compressed functional hierarchy in schizophrenia [3], yet no exploration of UHR-associated changes. This project aims to compare functional and structural cortical topography between UHR and control groups and within UHR groups with different transition outcomes. We expect to see group differences in functional hierarchy, larger effect sizes with functional measures compared to structural, and more severe functional changes in converters.
Methods:
We used data from Longitudinal Youth-at-Risk Study (LYRIKS) dataset [4] and retained 79 scans for the UHR positive (21.5±3.6 years, 52 males) and 43 scans (21.7±4.1 years, 20 males) for the control groups for further analysis. The Comprehensive Assessment of At Risk Mental State (CAARMS) was used to identify UHR individuals. All participants were followed up longitudinally for two years; UHR participants were assigned to either those who converted to psychosis during the course of the study (N=10, 19.3±3.3 years, 7 males) and those who did not (N=69, 21.8±3.6 years, 45 males).
The preprocessing pipeline was applied to resting state fMRI data following previous work [1]. Functional connectivity (FC) matrices were computed among 400 regions using Pearson's correlation [5]. We derived individual-level gradients from extracted FC matrices using Brainspace Toolbox [6]. For the top 2 gradients, values from 400 regions were averaged into 7 networks using a functional parcellation [7].
To extract reliable morphology estimates, images were automatically processed using FreeSurfer [8]. Based on the individual surface and volume templates, five structural features (surface area, cortical thickness, gray matter volume, Gaussian curvature, and mean curvature) were extracted. The morphometric similarity matrix was constructed for each individual and used as input to calculate morphometric similarity gradient following previous work [9].
To investigate structural and functional hierarchy of control and UHR subjects, we performed a series of t-test to see whether specific network gradients differed across groups. To study whether brain hierarchy correlated with disease severity, we calculated the Spearman correlation between gradients and total CAARMS score, which measures the severity of the symptoms. All analyses were controlled for age, sex, handedness, ethnicity.
The preprocessing pipeline was applied to resting state fMRI data following previous work [1]. Functional connectivity (FC) matrices were computed among 400 regions using Pearson's correlation [5]. We derived individual-level gradients from extracted FC matrices using Brainspace Toolbox [6]. For the top 2 gradients, values from 400 regions were averaged into 7 networks using a functional parcellation [7].
To extract reliable morphology estimates, images were automatically processed using FreeSurfer [8]. Based on the individual surface and volume templates, five structural features (surface area, cortical thickness, gray matter volume, Gaussian curvature, and mean curvature) were extracted. The morphometric similarity matrix was constructed for each individual and used as input to calculate morphometric similarity gradient following previous work [9].
To investigate structural and functional hierarchy of control and UHR subjects, we performed a series of t-test to see whether specific network gradients differed across groups. To study whether brain hierarchy correlated with disease severity, we calculated the Spearman correlation between gradients and total CAARMS score, which measures the severity of the symptoms. All analyses were controlled for age, sex, handedness, ethnicity.
Results:
The top two functional gradients separate the visual-somatomotor regions as well as the unimodal-transmodal regions (Fig. 1A). There was no difference between the two groups across all networks (FDR-corrected p-value>0.05; Fig. 1B, 1C). The same comparison between the control and the UHR positive population, now separated into the nonconverters and converters, returned no difference across groups. There was also no significant correlation between CAARMS score and gradients (FDR-corrected p-value>0.05).
The top two morphometric gradients correspond to the top two gradients reported in previous study [9] (Fig. 2A). Similarly, we observed no difference between controls and UHR participants across all networks (FDR-corrected p-value>0.05; Fig. 2B, 2C). There was also no difference in across conversion status or disease severity.
The top two morphometric gradients correspond to the top two gradients reported in previous study [9] (Fig. 2A). Similarly, we observed no difference between controls and UHR participants across all networks (FDR-corrected p-value>0.05; Fig. 2B, 2C). There was also no difference in across conversion status or disease severity.
Conclusions:
Surprisingly, both functional and morphometric gradients didn't differentiate the control and UHR positive groups or predict conversion status or disease severity. The small sample size, particularly for converters, may have contributed to this. It is also plausible that disruptions in cortical hierarchy emerge later in advanced disease stages. Larger studies are needed to confirm changes in hierarchy and identify the specific point in disease progression when they manifest.
Disorders of the Nervous System:
Psychiatric (eg. Depression, Anxiety, Schizophrenia) 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural) 2
Keywords:
FUNCTIONAL MRI
STRUCTURAL MRI
Other - psychosis
1|2Indicates the priority used for review