Multiple lines of research has shown that understanding the brain will require a global view on its dynamics. Different sensory and higher brain areas communicate continually with each other, modulating the excitability and function of connected regions.

This interplay of different brain areas is, unfortunately, compromised in many diseases such as depression, stroke, Alzheimers disease (AD), multiple sclerosis (MS), or epilepsy. Overall, brain disorders are a huge problem, causing suffering on a large scale and costing the society about 1000bn per year in Europe alone. In many cases, relief from brain disorders can be obtained from medication, surgery, or physical/cognitive therapy, but for many patients, these methods are either ineffective, too costly or accompanied with undesired side effects.

Recently, a new concept has been added to the arsenal of therapeutic methods: neuromodulation. The most effective forms of neuromodulation are electroconvulsive therapy (ECT) and deep brain stimulation (DBS), but good results have been demonstrated and regional governmental acceptance and insurance coverage have been obtained also for transcranial magnetic stimulation (TMS). However, current TMS technology is limited in that the locus (position) of stimulation can be changed only slowly; therefore, clinical neuromodulation paradigms almost always are limited to targeting the stimuli to only a single brain site, such as the dorsolateral prefrontal cortex in the case of medication-resistant depression, rather than targeting multiple nodes of a disordered cortical network.

Our goal in the ConnectToBrain project is to develop and prepare for wide clinical use of a non-invasive brain stimulation technique to stimulate brain networks in a feedback-controlled way.

By activating the neuronal network with TMS at suitable cortical locations at suitable time instants depending on the instantaneous brain state, as measured by electroencephalography (EEG), we can couple to the neurodynamics of the brain.

Our hypothesis is that with clever algorithms, we will be able to control brain dynamics in such a way that we can guide it towards better mood (in the case of depression), towards better connectivity (stroke, MS, or AD), or towards diminished inter-area excitability (epilepsy). For this purpose, the European Research Council (ERC) awarded a Synergy Grant of 10m to the ConnectToBrain project for the period 20192025.

1) Multilocus TMS (mTMS) coil array that will cover most of the cortical mantle and will allow accurate real-time control of the location, direction, intensity and timing of millisecond-time-scale sequences of brain-activating pulses,

2) Real-time analysis of brain activity and connectivity based on high-density EEG for brain-state-dependent and closed-loop stimulation and for adaptive optimisation of treatment effects by machine learning,

3) Application of these technologies and methodologies in scientific studies and brain therapy. If all these steps are successful and the full therapeutic potential of ConnectToBrain will be realised, the new methodology may eventually lead to a substantial reduction of the cost of brain disorders.

Fig. 1 is an artists view of the multilocus magnetic stimulator of the future. The white helmet-shaped structure (here, consisting of two parts, for the two hemispheres) will contain a large number of partially overlapping coils, which would be activated with variable relative intensities synchronously to produce the desired three-dimensional form of the electromagnetic field. By changing the relative intensities, the hotspot (stimulated location) in the brain can be varied at millisecond precision and millimeter accuracy, allowing one to stimulate either separate sites in rapid succession or multiple nodes of the network simultaneously.

The changing magnetic field, penetrating the scalp and skull freely, induces in the brain electric currents that give rise to neuronal signals. The brain signals, called action potentials, although triggered artificially, are perfectly normal and therefore pose minimal risk to the patient. When the hand area of the motor cortex is stimulated, much to the surprise of a first-time subject, a finger in the contralateral hand can move as a result. By stimulating different nodes of brain networks, this technique can be used to strengthen brain connections that have been weakened by brain disorders such as depression, stroke, MS, or AD. The challenge is to develop stimulation sequences that are optimal for the therapeutic benefit.

A key technology to be developed in the project consists of devoted software algorithms that will control the stimulation sequences (see Fig. 2). A typical algorithm will start at pre-selected sites in the brain and then, depending on observed effects on brain activity and brain connectivity, it will automatically and rapidly move the stimulation target in order to approach brain-modulation or therapeutic goals. Brain activity and brain connectivity will be derived in real-time using computationally efficient strategies that will allow predicting the evolution of brain states with millisecond resolution.

To help brain activity and connectivity estimation, information from different neuroimaging modalities in the same subjects or from neuroimaging databases will act as priors. Approaches based on artificial intelligence (AI), for example in a machine learning framework, will be used to automatically derive relevant features that capture changes in the brain state induced by the stimulation, and will transform these features into new driving inputs for the stimulator module.

Conventional single-locus open-loop non-adaptive therapeutic brain stimulation shows serious limitations with respect to effect size, interindividual variability of therapeutic response and high proportions of non-responders. The three synergistic groups will implement the technological developments of mTMS in combination with high-density EEG (hdEEG) and the data analysis algorithms for closed-loop neuromodulation in translational experiments. First, studies will be conducted in healthy subjects to demonstrate feasibility, safety, and neurophysiological and behavioural effects of closed-loop hdEEGmTMS to modify excitability and connectivity in motor and working-memory brain networks. Later, in patient groups with motor stroke or AD, we aim at demonstrating therapeutic utility in their dysfunctional, respectively, motor or working-memory networks. At the end of the funding period of ConnectToBrain, we expect to have new technology capable of correcting dysfunctional brain networks in a variety of neurological and psychiatric disorders with far better therapeutic efficacy than hitherto possible.

Professor Risto J IlmoniemiAalto Universityristo.ilmoniemi@aalto.fi

Professor Emeritus Gian Luca RomaniG. DAnnunzio University of Chieti-Pescaraglromani@itab.unich.it

Professor Ulf ZiemannUniversity of Tbingenulf.ziemann@uni-tuebingen.de

Disclaimer: This article is featured in the December issue of SciTech Europa Quarterly.

Visit link:

ConnectToBrain: The dynamics of the brain - SciTech Europa