Dynamic sensitivity analysis: Defining personalised strategies to drive brain state transitions via whole brain modelling

Vohryzek, Jakub; Cabral, Joana; Castaldo, Francesca; Perl, Yonatan Sanz; Lord, Louis-David; Fernandes, Henrique M.; Kringelbach, Morten L.; Deco, Gustavo

doi:10.1016/j.csbj.2022.11.060

Cited by 25 publications

(26 citation statements)

References 114 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unraveling the mechanisms of different clinical interventions, and designing suitable stimulation protocols are trending topics in computational and clinical neuroscience, which draw upon methods from perturbation analysis (Deco et al 2019;Mana et al 2022;Sanz Perl et al 2022;Vohryzek et al 2022) and control theory (Gu et al 2017;Gu et al 2022;Kamiya et al 2023;Srivastava et al 2020;Yang et al 2021) Notably, optimizing the stimulation locations(s) is usually performed using exhaustive search, which is computationally expensive (Deco et al 2019;Mana et al 2022;Vohryzek et al 2022). Recent work suggests that the nodes' controllability 27 at rest can help to identify the optimal stimulation locations (Yang et al 2021).…”

Section: Discussionmentioning

confidence: 99%

“…Despite technical variations, the conceptual procedure remains the same: model the brain as a dynamical system; fit this model to the spatiotemporal features of the brain in the unperturbed state; add a stimulation term to the model (or modulate some model parameters) to trigger transition to a new brain state; finally, optimize the location and profile of the stimulation or modulation for (in silico) transition to the new state. Depending on the study, the source/target brain states may correspond to the profiles of transient brain dynamics in normal/pathological(Mana et al 2022;Vohryzek et al 2022) or sleep/awake(Deco et al 2019) scenarios; alternatively, the states may be defined as the spatiotemporal patterns of rest/task(Kamiya et al 2023) rest/stimulation(Yang et al 2021) conditions.…”

mentioning

confidence: 99%

See 1 more Smart Citation

A new causal centrality measure reveals the prominent role of subcortical structures in the causal architecture of the extended default mode network

Zarghami

2023

Preprint

View full text Add to dashboard Cite

Network representation has been a groundbreaking concept for understanding the behavior of complex systems in social sciences, biology, neuroscience, and beyond. Network science is mathematically founded on graph theory, where nodal importance is gauged using measures of centrality. Notably, recent work suggests that the topological centrality of a node should not be over-interpreted as its dynamical or causal importance in the network. Hence, identifying the influential nodes in dynamic causal models (DCM) remains an open question. This paper introduces causal centrality for DCM, a dynamics-sensitive and causally-founded centrality measure based on the notion of intervention in graphical models. Operationally, this measure simplifies to an identifiable expression using Bayesian model reduction. As a proof of concept, the average DCM of the extended default mode network (eDMN) was computed in 74 healthy subjects. Next, causal centralities of different regions were computed for this causal graph, and compared against major graph-theoretical centralities. The results showed that the subcortical structures of the eDMN are more causally central than the cortical regions, even though the (dynamics-free) graph-theoretical centralities unanimously favor the latter. Importantly, model comparison revealed that only the pattern of causal centrality was causally relevant. These results are consistent with the crucial role of the subcortical structures in the neuromodulatory systems of the brain, and highlight their contribution to the organization of large-scale networks. Potential applications of causal centrality - to study other neurotypical and pathological functional networks - are discussed, and some future lines of research are outlined.

show abstract

Section: Discussionmentioning

confidence: 99%

mentioning

confidence: 99%

A new causal centrality measure reveals the prominent role of subcortical structures in the causal architecture of the extended default mode network

Zarghami

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…These initiatives have generated a lot of valuable data to study, diagnose and analyse the disease progression 1 . Yet, a more mechanistic interpretation is required both for understanding the disease itself 3 and for predicting the outcomes of potential pharmacological interventions, noninvasive electrical stimulations and cognitive training treatments [4][5][6] .…”

Section: Introductionmentioning

confidence: 99%

“…Conceptually, we can consider every neurodegenerative or psychiatric condition, such as Alzheimer's, Schizophrenia and Autism, to be depicted by a condition-specific brain state with altered spatio-temporal dynamics compared to that of a neurotypical (healthy) brain state 5 . This opens avenues towards the diagnosis, prognosis, and potential treatment intervention paradigms 7 .…”

Section: Introductionmentioning

confidence: 99%

“…This description in turn serves for mechanistic scenarios where in-silico models use information at lower scales (microscopic or mesoscopic) to causally explain the observed changes at the large-scale brain state level 10 . Furthermore, it can serve for stimulation paradigms where mechanistic explanations are relevant in suggesting adequate protocols [4][5][6] .…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Design of effective personalised perturbation strategies for enhancing cognitive intervention in Alzheimer’s disease

Vohryzek

Cabral

Perl

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

One of the potential and promising adjuvant therapies for Alzheimer’s disease is that of non-invasive transcranial neurostimulation to potentiate cognitive training interventions. Conceptually, this is achieved by driving brain dynamics towards an optimal state for an effective facilitation of cognitive training interventions. However, current neurostimulation protocols rely on experimental trial-and-error approaches that result in variability of symptom improvements and suboptimal progress. Here, we leveraged whole-brain computational modelling by assessing the regional susceptibility towards optimal brain dynamics from Alzheimer’s disease. In practice, we followed the three-part concept of Dynamic Sensitivity Analysis by first understanding empirical differences between healthy controls and patients with mild cognitive impairment and mild dementia due to Alzheimer’s Disease; secondly, by building computational models for all individuals in the mild cognitive impairment and mild dementia cohorts; and thirdly, by perturbing brain regions and assessing the impact on the recovery of brain dynamics to the healthy state (here defined in functional terms, summarised by a measure of metastability for the healthy group). By doing so, we show the importance of key regions, along the anterior-posterior medial line, in driving in-silico improvement of mild dementia and mild cognitive impairment groups. Moreover, this subset consists mainly of regions with high structural nodal degree. Overall, this in-silico perturbational approach could inform the design of stimulation strategies for re-establishing healthy brain dynamics, putatively facilitating effective cognitive interventions targeting the cognitive decline in Alzheimer’s disease.

show abstract

Virtual external stimulation promotes the transformation of the brain state from early mild cognitive impairment to health

Wang,

Zhao

et al. 2023

Brain-X

View full text Add to dashboard Cite

Neurostimulation has emerged as a potential remedy for early mild cognitive impairment (EMCI). However, further exploration is needed on how external stimulation of brain regions promotes the transition of the brain state from EMCI to health and the selection of target locations. In this study, a functional magnetic resonance imaging dataset was used to evaluate the brain states of healthy individuals and patients with EMCI to explore the probabilistic metastable substate space, identifying abnormal manifestations of EMCI. Stimulation targets were then identified and stimulated to achieve complete controllability of the effective connection network for EMCI. A whole‐brain model successfully fitted the brain state of the patients with EMCI based on diffusion tensor imaging data. Based on this whole‐brain model, stimulation of the hippocampus, medial frontal gyrus, suboccipital gyrus, and fusiform gyrus can promote the transformation of the brain state from EMCI to health. The findings reveal the underlying brain mechanisms of cognitive decline in patients with EMCI and the stimulation targets of the neural mechanisms of EMCI restoration, which could help in designing more effective therapeutic interventions for EMCI.

show abstract

Dynamic sensitivity analysis: Defining personalised strategies to drive brain state transitions via whole brain modelling

Cited by 25 publications

References 114 publications

A new causal centrality measure reveals the prominent role of subcortical structures in the causal architecture of the extended default mode network

A new causal centrality measure reveals the prominent role of subcortical structures in the causal architecture of the extended default mode network

Design of effective personalised perturbation strategies for enhancing cognitive intervention in Alzheimer’s disease

Virtual external stimulation promotes the transformation of the brain state from early mild cognitive impairment to health

Contact Info

Product

Resources

About