Temporal Irreversibility of Large-Scale Brain Dynamics in Alzheimer’s Disease

Abstract

Healthy brain dynamics can be understood as the emergence of a complex system far from thermodynamic equilibrium. Brain dynamics are temporally irreversible and thus establish a preferred direction in time (i.e., arrow of time). However, little is known about how the time-reversal symmetry of spontaneous brain activity is affected by Alzheimer's disease (AD). We hypothesized that the level of irreversibility would be compromised in AD, signaling a fundamental shift in the collective properties of brain activity towards equilibrium dynamics. We investigated the irreversibility from resting-state fMRI and EEG data in male and female human patients with AD and elderly healthy control subjects (HC). We quantified the level of irreversibility and, thus, proximity to non-equilibrium dynamics by comparing forward and backward timeseries through time-shifted correlations. AD was associated with a breakdown of temporal irreversibility at the global, local, and network levels and at multiple oscillatory frequency bands. At the local level, temporoparietal and frontal regions were affected by AD. The limbic, frontoparietal, default mode, and salience networks were the most compromised at the network level. The temporal reversibility was associated with cognitive decline in AD and grey matter volume in HC. The irreversibility of brain dynamics provided higher accuracy and more distinctive information than classical neurocognitive measures when differentiating AD from controls. Findings were validated using an out-of-sample cohort. Present results offer new evidence regarding pathophysiological links between the entropy generation rate of brain dynamics and the clinical presentation of AD, opening new avenues for dementia characterization at different levels.SIGNIFICANCE STATEMENT:By assessing the irreversibility of large-scale dynamics across multiple brain signals, we provide a precise signature capable of distinguishing Alzheimer's disease (AD) at the global, local, and network levels and different oscillatory regimes. Irreversibility of limbic, frontoparietal, default mode, and salience networks was the most compromised by AD in comparison to more sensory-motor networks. Moreover, the time-irreversibility properties associated with cognitive decline and atrophy outperformed and complemented classical neurocognitive markers of AD in predictive classification performance. Findings were generalized and replicated with an out-of-sample validation procedure. We provide novel multilevel evidence of reduced irreversibility in AD brain dynamics that have the potential to open new avenues for understating neurodegeneration in terms of the temporal asymmetry of brain dynamics.

Model-based whole-brain perturbational landscape of neurodegenerative diseases

“…Alzheimer’s disease (AD) and behavioral variant frontotemporal dementia (bvFTD) are among the most prevalent neurodegenerative diseases, each linked to specific pathophysiology and to highly heterogeneous and atypical manifestation, hindering detection and diagnosis ( Musa et al, 2020 ; Parra et al, 2018 ). Neuroimaging methods such as positron emission tomography (PET) ( Engler et al, 2008 ; Foster et al, 2007 ; Jack et al, 2018 ; Mosconi et al, 2008 ; Nordberg, 2004 ), functional magnetic resonance imaging (fMRI) ( Hafkemeijer et al, 2015 ; Jalilianhasanpour et al, 2019 ; Moguilner et al, 2021 ), and electroencephalography (EEG) Dottori et al, 2017 ; Lindau et al, 2003 ; Nishida et al, 2013 ; Nishida et al, 2011 ; Cruzat et al, 2023 have been employed extensively to profile these diseases based on their underlying whole-brain activity patterns, and to develop automated classifiers capable of assisting clinical decision-making ( Kim et al, 2019 ; Moguilner et al, 2021 ; Herzog et al, 2022 ). In spite of these advances, the limited knowledge of the large-scale network mechanisms of neurodegenerative dementias hinders the development of interventions oriented to improve brain health and subjective well-being.…”

Navigating Pubertal Goldilocks: The Optimal Pace for Hierarchical Brain Organization

“…Harnessing the principles of thermodynamics, we quantified hierarchy by computing the level of irreversibility in brain signals during resting state conditions. Specifically, we captured the level of irreversibility through entropy production 69,71,74,76,77 as computed through a sophisticated clustering strategy. Note that resting states cover a rich range of the dynamical repertoire of the brain and thus those are ideal to characterise the most general and wide variety of interaction across the whole brain, i.e., the underlying hierarchical orchestration.…”

Multilevel irreversibility reveals higher-order organisation of non-equilibrium interactions in human brain dynamics

“…Then the data was converted into Statistical Parametric Mapping (SPM) formatting and further pre-processed in MATLAB (MathWorks, Natick, MA, USA) using in-house-built codes (LBPD, available at https://github.com/leonardob92/LBPD-1.0.git) and the Oxford Centre for Human Brain Activity (OHBA) Software Library (OSL) (available at https://ohba-analysis.github.io/osl-docs/) [38]. OSL is freely available software that builds on the Fieldtrip [29], FSL [39], and SPM [12] toolboxes. Next the continuous MEG data was visually inspected and large artefacts were removed.…”

“…This fundamental observation has prompted the development of a range of techniques to provide a measure for the degree of non-equilibrium in neuroimaging time-series recorded in different conditions [5–10]. These measures have shown that the degree of non-equilibrium is elevated during cognitive tasks [4–7] whilst reduced in both impairments of consciousness [11], sleep [10] and Alzheimer’s disease [12] indicating that non-equilibrium may be a key signature of healthy consciousness and cognition in the brain [13]. Despite this, current methods are restricted to aggregate measures of non-equilibrium.…”