The colors of our brain: an integrated approach for dimensionality reduction and explainability in fMRI through color coding (i-ECO)

Abstract

Several systematic reviews have highlighted the role of multiple sources in the investigation of psychiatric illness. For what concerns fMRI, the focus of recent literature preferentially lies on three lines of research, namely: functional connectivity, network analysis and spectral analysis. Data was gathered from the UCLA Consortium for Neuropsychiatric Phenomics. The sample was composed by 130 neurotypicals, 50 participants diagnosed with Schizophrenia, 49 with Bipolar disorder and 43 with ADHD. Single fMRI scans were reduced in their dimensionality by a novel method (i-ECO) averaging results per Region of Interest and through an additive color method (RGB): local connectivity values (Regional Homogeneity), network centrality measures (Eigenvector Centrality), spectral dimensions (fractional Amplitude of Low-Frequency Fluctuations). Average images per diagnostic group were plotted and described. The discriminative power of this novel method for visualizing and analyzing fMRI results in an integrative manner was explored through the usage of convolutional neural networks. The new methodology of i-ECO showed between-groups differences that could be easily appreciated by the human eye. The precision-recall Area Under the Curve (PR-AUC) of our models was > 84.5% for each diagnostic group as evaluated on the test-set – 80/20 split. In conclusion, this study provides evidence for an integrative and easy-to-understand approach in the analysis and visualization of fMRI results. A high discriminative power for psychiatric conditions was reached. This proof-of-work study may serve to investigate further developments over more extensive datasets covering a wider range of psychiatric diagnoses.

Progressive Voxel‐Wise Homotopic Connectivity from childhood to adulthood: Age‐related functional asymmetry in resting‐state functional magnetic resonance imaging

“…A small effect for age was previously shown for the resting‐state activity of the DMN—in a sample of adults (Bluhm et al., 2008)—while the current study indicates medium age effects in the same region from 7 to 18 years of age. The crucial role of DMN in typical neurodevelopment was also shown to be observable by a variety of dimensions, as network connectivity and centrality (Davey et al., 2019; Gu et al., 2015; Sato et al., 2014; Tarchi et al., 2021). DMN shows increasing intra‐ and internetwork connectivity during childhood (Gu et al., 2015) and increasing centrality through early adolescence (Sato et al., 2014).…”

Centrality and interhemispheric coordination are related to different clinical/behavioral factors in attention deficit/hyperactivity disorder: a resting-state fMRI study

“…Attention Deficit/Hyperactivity disorder (ADHD) is characterized by symptoms presenting in a heterogeneous manner across individuals, including attention deficits, impulsivity, and hyper-activity (American Psychiatric Association, 2013 ). Functional Magnetic Resonance Imaging (fMRI) proved to be a powerful tool for exploring the neurobiological correlates of ADHD symptoms and behaviors (Damiani et al, 2020 ; Iravani et al, 2021 ; Qian et al, 2018 ; Rosch et al, 2018 ; Silva et al, 2021 ; Tarchi et al, 2021 ). Specifically, fMRI highlighted the importance of how each region is functionally connected to the rest of the brain.…”

“…This conceptualization provides an efficient and simple instrument to better explore the complex functional organization of the brain, also known as the functional connectome (Iturria-Medina et al, 2008 ; Sporns, 2006 ; Sporns et al, 2005 , 2007 ). Centrality measurements proved to have the ability to capture intrinsic features of the human functional connectome in both neurotypicals (Achard et al, 2006 ; He et al, 2009 ; Sporns et al, 2007 ; Tarchi et al, 2021 ; Zuo et al, 2012 ), and individuals with neuropsychiatric disorders (Reinelt et al, 2019 ; Seidel et al, 2020 ), including ADHD (M. Zhou et al, 2019 ). However, the available evidence showed both increased and decreased centrality scores in ADHD compared to neurotypical controls, in particular for the superior Temporal lobes and the middle/inferior Occipital lobes (Di Martino et al, 2013 ; Hong et al, 2017 ; Tarchi et al, 2021 ; Zhou et al, 2019 ).…”

“…Centrality measurements proved to have the ability to capture intrinsic features of the human functional connectome in both neurotypicals (Achard et al, 2006 ; He et al, 2009 ; Sporns et al, 2007 ; Tarchi et al, 2021 ; Zuo et al, 2012 ), and individuals with neuropsychiatric disorders (Reinelt et al, 2019 ; Seidel et al, 2020 ), including ADHD (M. Zhou et al, 2019 ). However, the available evidence showed both increased and decreased centrality scores in ADHD compared to neurotypical controls, in particular for the superior Temporal lobes and the middle/inferior Occipital lobes (Di Martino et al, 2013 ; Hong et al, 2017 ; Tarchi et al, 2021 ; Zhou et al, 2019 ). Analyses of the age contribution to centrality measurements in ADHD indicated a role for development in moderating the Resting-State fMRI activity in the middle Temporal cortex (Hong et al, 2017 ), with additional reports of transient alterations during development among patients with ADHD (Damiani et al, 2020 ; Hong et al, 2017 ).…”

“…Although previous studies focused on voxel-wise differences between ADHD and neurotypicals, the current work also adopted a network-based approach to provide novel insights on EC/VMHC. This approach allows to clearly visualize the relationship between neuroimaging and clinical findings (Tarchi et al, 2021 ), and to improve their replicability (Nickerson, 2018 ). The secondary aims of this study were i) to evaluate potential differences between neurotypicals and patients with ADHD in EC and VMHC at the network level.…”