Using computational models and clinical data from depth electrode implants to evaluate active probing paradigms in epilepsy

• Ano de defesa: 2021
• Autor(a) principal: Vinícius Rezende Carvalho
• Orientador(a): Eduardo Mazoni Andrade Marçal Mendes
• Banca de defesa: Antonio Carlos Roque da Silva Filho, Jean Faber Ferreira de Abreu, Vinícius Rosa Cota, Leonardo Antônio Borges Tôrres
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Universidade Federal de Minas Gerais
• Programa de Pós-Graduação: Programa de Pós-Graduação em Engenharia Elétrica
• Departamento: ENG - DEPARTAMENTO DE ENGENHARIA ELÉTRICA
• País: Brasil
• Palavras-chave em Português: Epilepsy; Active probing; Critical slowing down; Computational neuroscience; Seizure prediction; SEEG
• Link de acesso: http://hdl.handle.net/1843/40181; https://orcid.org/0000-0003-1821-1634

Recurrent and spontaneous seizures are the hallmark of epilepsy, a neurological disease that affects around 1% of the world population and has higher incidence in developing countries. In general, seizure (or ictal) events are relatively easily identified due to their hypersynchronous and hyperexcitable nature, but the underlying neuronal mechanisms that give rise to such events are still unclear. One of the main ways of elucidating this involves the extraction of relevant features from electroencephalographic signals that are related to abnormal neuronal functioning. One promising strategy to highlight such features is the use of active probing paradigms. That is, the use of probing stimuli that would reveal information from a dynamical system, that would be otherwise inaccessible (or difficult to detect) through passive observation. This work evaluates the use of such probing strategies in two different contexts. In the first part of this thesis, a modified version of a well-known neural mass model is used to test how low-frequency probing stimuli enhance the detection of approaching seizures. We show that, as parameters are shifted to bring the model towards seizure activity, features extracted from the simulated activity reflect these changes when probing stimuli are used. Furthermore, the changes revealed by the stimuli appear to reflect the phenomenon of critical slowing down, where increased recovery times from perturbations may signal the loss of a systems’ resilience and are common hallmarks of an impending critical transition. Incorporating this strategy into early-warning systems may ultimately enable closing the loop for targeted seizure-controlling interventions. In the second part of the work, an auditory task was used to evaluate sound processing in epilepsy patients undergoing presurgical monitoring with depth electrode implants. It was hypothesized that regions involved in the onset of seizures would display abnormal auditory steady-state responses (ASSR) to amplitude-modulated sounds. This could be in the form of abnormal hypersynchronous responses in regions not usually involved in primary auditory processing, or in impaired responses due to central auditory dysfunctions caused by epileptogenic circuits. Results show that ASSRs were rarely elicited in electrode contacts within seizure onset zones (SOZ), contradicting the first hypothesis. Accurately evaluating the second hypothesis (of impaired ASSRs within SOZ) was not possible due to the highly heterogeneous nature of the data and implant schemes from all participants. This work showed the potential of low-frequency probing stimuli to highlight features and reveal underlying dynamics involved in the transition from interictal to ictal activity. However, the use of ASSRs for accurately identifying putative epileptogenic regions was limited. These responses are still promising for other diagnostics, such as identifying lateralization or seizure susceptibility, as may other types of stimuli be useful for identifying epileptogenic regions. Computational and experimental models will be crucial for defining the best strategies and parameters of active probing approaches for the identification of abnormal and epileptic activity.