| dc.description.abstract | Post-traumatic stress disorder (PTSD) is a psychiatric disorder that may develop when an individual ex- periences or witnesses a traumatic stressful event. Although trauma-focused psychotherapy is a common approach for treatment, a significant proportion of patients continue to experience symptoms even after therapy. To enhance treatment response rates, it is important to increase the understanding of the neurobi- ological factors that may predict treatment response.
Recently, there has been a growing interest in characterizing neural activity in terms of order and disorder. It is theorized that the brain operating close to the border between order and disorder presents optimized information processing. The concepts of ’criticality’ and ’entropy’ provide analytical frameworks for studying these phenomena.
This thesis, conducted on behalf of the Expertisecentrum MGGZ of the Dutch Military, aims to investigate whether the criticality and entropy concepts are useful for explaining differences in treatment response. For this study, a set of resting-state functional magnetic resonance imaging (fMRI) data is used, acquired from Dutch soldiers diagnosed with PTSD before undergoing psychotherapy.
The first part of the thesis focuses on criticality, relying on the Pairwise Maximum Entropy Model. Due to constraints in data sampling, an archetype model is derived from the concatenated timeseries of the participants. Two approaches are presented, involving the inference of a ’personal’ system temperature, analogous to the Ising model, and drawing phase diagrams inspired by the Sherrington-Kirkpatrick model. These methodologies aim to ’personalize’ the archetype model and serve as metrics to compare the distance from criticality.
In the second part of the thesis, the concept of entropy is explored using the Permutation Fuzzy Entropy algorithm. This algorithm is chosen for its reliability in estimating Shannon entropy of neural signals in fMRI data. | |
