Investigating the impact of novel personalized neurostimulation strategies to promote recovery after brain lesions

Nowadays, non-communicable diseases represents the 80\% of the top ten causes of death worldwide. Among these disorders, stroke is the second leading cause of death and disability. Current assistive technology is limited and only a minority of survivors are able to achieve functional independence in simple activities of daily living. Thus, understanding the impact of a lesion on the brain networks and promoting functional and behavioral recovery has become a global priority in healthcare. The standard-of-care for post-stroke rehabilitation is physical therapy but its effects are often limited or incomplete. To this end, research is exploring innovative approaches, such as the ‘electroceutical’ one. Recent studies have shown how brain stimulation could enhance functions after stroke in both animals and patients. These studies have used electrical stimulation patterns that are ‘adapted’ (i.e., personalized) to the intrinsic dynamics of the networks under examination, primarily through two different modalities, i.e., open loop and closed loop. By using personalized stimulation is possible to: (i) promote both plasticity and motor recovery, as in the case of stroke or traumatic brain injury –TBI; (ii) block the pathological activity and promote the physiological one within the damaged tissue, as in the case of Parkinson's disease or epilepsy; (iii) ‘entrain’ the network under examination and facilitate motor responses. Most of these studies focused mainly on motor recovery but not on the evaluation of functional reorganization and on the possible changes in electrophysiological activity of the involved networks. Recent results demonstrated the capabilities of Activity-Dependent stimulation (i.e., ADS), a closed loop approach, to better entrain network activity with respect to standard open loop random stimulation. Currently, at the clinical level, stimulation-based therapy relies on standardized protocols with not always convincing results. A possible explanation for these inconsistent outcomes is the lack of a stimulation protocol ‘tailored’ on the intrinsic dynamics of the target system. \textit{Objective}. This study aims at going beyond the state of the art and improving current neuromodulation techniques by customizing electrical stimulation on the electrophysiological features of the individual to be treated. Specifically, we aim at investigating possible changes in electrophysiological activity induced by a focal lesion in animal models in-vivo and the subsequent effect of a personalized, open loop stimulation protocol, on both behavior and functions. Thus, this PhD work has the following three main sub-goals: 1) to investigate the effects of an ischemic lesion in anesthetized rats; 2) to design and characterize the short-term effects of novel personalized neurostimulation techniques on neural activity in anesthetized rats; 3) to investigate the effects of a prolonged personalized stimulation protocol on both neural activity and behavior of injured awake animals. \textit{Approach}. We used Long Evans rats implanted in the Rostral Forelimb Area (i.e., RFA, equivalent of the pre-motor cortex in the rat) and in the Somatosensory cortex (i.e., S1). An ischemic lesion was performed in the equivalent of motor cortex (i.e., CFA) by means of a local injection of Endothelin-1. Firstly, we evaluated the effect of the lesion on anesthetized animals in which we tested novel personalised stimulation protocols that we developed by following an open-loop paradigm. The personalization was achieved by designing a stimulation pattern, which reproduced the intrinsic dynamics of a target area. To design the personalized stimulation, we processed the healthy spontaneous activity of a selected channel in the rostral forelimb area (RFA) with one of the following methods: i) Exponential Stimulation: we designed a new spike sequence with some features of the network of reference, according to a pre-defined/mathematical distribution, e.g. exponential distribution. ii) Repeated Stimulation: by following a ‘replay’ and ‘repeat’ strategy, we created a stimulation pattern based on what was recorded during spontaneous activity; iii) Shuffled Stimulation: we generated a new spike sequence according to a data-driven ISI, based on data acquired from the network under experiment. \textit{Main results}. We found that the lesion had a clear-cut effect on decreasing the spontaneous activity at the two monitored locations. In anesthetized animals, the personalized open-loop (i.e., repeated stimulation and shuffled stimulation) protocols showed a tendency to increase the level of spiking activity. These results are in line with the effects induced by a closed loop paradigm named Activity Dependent Stimulation, which demonstrated to be effective in promoting the post-lesion recovery. We can then conclude that non-periodic, tailored stimulation might be the key to restore the firing patterns in the damaged brain. Moreover, we also found that repeated stimulation treatment enhanced motor recovery in behaving animals. \textit{Significance}. These results have the potential to lead novel neurostimulation techniques for treating neurological diseases, thus opening up the possibility of translating the personalized electroceutical therapy to the clinical level.