Investigating Disruptions in Information Flow due to Sickle Cell Disease using Granger Causality
Poster No:
309
Submission Type:
Abstract Submission
Authors:
Nahom Mossazghi1, Nadim Farhat2, Tales Santini3, Olubusola Oluwole3, Enrico Novelli3, Tamer Ibrahim3, Sossena Wood1
Institutions:
1Carnegie Mellon University, Pittsburgh, PA, 2University of Pittsburgh, Pittsburrgh, PA, 3University of Pittsburgh, Pittsburgh, PA
First Author:
Co-Author(s):
Introduction:
Sickle cell disease (SCD) is an inherited blood disorder characterized by a mutation in the gene encoding for the beta chain of hemoglobin1,2. Patients with SCD experience various complications, including a decline in executive functions. Neuroimaging studies have revealed SCD-related structural differences, yet their influence on functional connectivity remains unclear3. Reduced activity in the Executive Control Network (ECN) has been linked to increased pain processing, which diverts resources from the ECN in adults with SCD compared to Healthy Controls (HC)4. Our study applied Granger causality analysis to investigate dynamic interactions among brain regions within the ECN and other resting-state networks. Building upon previous research, we hypothesized that (1) adults with SCD would exhibit lower information flow between brain regions associated with ECNs compared to HC, and (2) SCD patients would demonstrate reduced directional influence, measured by net information flow, compared to HC.
Methods:
Structural and functional MRI data were obtained from 19 adults, including steady-state patients with SCD (n=9 (8 HBSC, 1 HbSβ+thalassemia), mean age=32.3+/- 8.2 years) and matched HC (n=10, mean age=36.1+/- 8.1 years) using a 7T scanner (MAGNETOM, Siemens). The head coil consists of 16 transmit channels, and 32 receive channels and provides 5. Resting-state fMRI (rs-MRI) scans with 86 axial slices were acquired (TR/TE = 2500/20 ms, flip angle = 65°, voxel size= 1.50 mm iso, FOV = 222 × 222 mm, multiband factor = 2, slice thickness = 1.50 mm, acquisition time = 5:45 min). Preprocessing employed fMRIprep and time series extraction using the Schaefer 2018 atlas (Fig 1 (a))6,7. Information flow strength between brain regions was assessed via Granger F-values, optimizing lag time via Akaike Information Criterion (p < 0.05 significance). We calculated average F-values for brain area pairs and employed an independent t-test to compare HC and SCD. Additionally, we computed 'net' information flow by subtracting efferent from afferent F-values for each brain area. We examined its significance using the Wilcoxon test to detect directionally influence in information flow8.
Results:
HC demonstrated stronger information exchange between pairs of ROIs, with a global mean F-value of 3.02, while patients with SCD had a global mean of 2.23 (p-value < 0.001), Fig 1(b-d). Furthermore, Fig 1(e-f) shows that inter-hemispheric information flow (between the left (L) and right (R) hemispheres) was higher in HC than in SCD patients. The mean F-values for L → R were 3.32 and 2.09 for HC and SCD patients (p-value < 0.001), while the mean F-values for R → L were 3.15 and 2.28 for HC and SCD patients (p-value < 0.001) respectively. The magnitude of information exchange was higher in HC than in patients with SCD, as shown in Fig 2(a, c). However, the p-value analysis in Fig 2(b, d) indicates a balanced net F-value. Nevertheless, our analysis revealed some areas exhibit directional influence, with more instances found in patients with SCD than HC. We did not observe any significant differences in other resting state networks.
·Figure 1: Average information flow within the Executive Control Network (ECN) measured by F-Value.
·Figure 2. Net information flow map between brain areas in the ECN.
Conclusions:
Our preliminary results further extend previous reports in SCD, which found that decreased signal in effective connectivity between brain regions within the ECN4. Similar findings have been reported in mild cognitive impairment and Alzheimer's disease studies, indicating a relationship between decreased signal and abnormalities in effective connectivity within the ECN9,10. Our study faced several limitations, including the small sample size we analyzed and the inherent limitations of fMRI, such as its slower neural response interpretation and the potential influence of vasculature on the BOLD signal, despite its neural basis. Future analyses will delve deeper into understanding the effects of SCD on effective connectivity through task-based DSST fMRI studies.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Higher Cognitive Functions:
Executive Function, Cognitive Control and Decision Making 2
Keywords:
Computational Neuroscience
MRI
1|2Indicates the priority used for review
Provide references using author date format
2. Kavanagh, P. L., Fasipe, T. A. & Wun, T. Sickle Cell Disease: A Review. JAMA vol. 328 57–68 Preprint at https://doi.org/10.1001/jama.2022.10233 (2022).
3. Santini, T. et al. Analysis of hippocampal subfields in sickle cell disease using ultrahigh field MRI. Neuroimage Clin 30, (2021).
4. Case, M. et al. Characterization of functional brain activity and connectivity using EEG and fMRI in patients with sickle cell disease. Neuroimage Clin 14, 1–17 (2017).
5. Santini, T. et al. Improved 7 Tesla transmit field homogeneity with reduced electromagnetic power deposition using coupled Tic Tac Toe antennas. Sci Rep 11, (2021).
6. Esteban, O. et al. fMRIPrep: a robust preprocessing pipeline for functional MRI. Nat Methods 16, 111–116 (2019).
7. Schaefer, A. et al. Local-Global Parcellation of the Human Cerebral Cortex from Intrinsic Functional Connectivity MRI. Cerebral Cortex 28, 3095–3114 (2018).
8. Hutchison, R. M., Womelsdorf, T., Gati, J. S., Everling, S. & Menon, R. S. Resting-state networks show dynamic functional connectivity in awake humans and anesthetized macaques. Hum Brain Mapp 34, 2154–2177 (2013).
9. Zhao, Q., Lu, H., Metmer, H., Li, W. X. Y. & Lu, J. Evaluating functional connectivity of executive control network and frontoparietal network in Alzheimer’s disease. Brain Res 1678, 262–272 (2018).
10. Liu, Z. et al. Investigation of the effective connectivity of resting state networks in Alzheimer’s disease: A functional MRI study combining independent components analysis and multivariate Granger causality analysis. NMR Biomed 25, 1311–1320 (2012).