Αναζήτηση

Πλοήγηση

• Κοινότητες

Ο Χώρος μου

• Είσοδος

Λειτουργικές εναλλαγές εγκεφαλικών δικτύων σε ήπιες κρανιοεγκεφαλικές κακώσεις αξιοποιώντας ρεαλιστικά μοντέλα κεφαλής πεπερασμένων στοιχείων και ΗΕΓ/ΜΕΓ

Politof Konstantinos

Πλήρης Εγγραφή

---

Περίληψη

Motivation: Mild traumatic brain injury (mTBI) induces subtle neural changes that are often undetected by traditional neuroimaging techniques. Due to variability in findings across studies using EEG and MEG functional modalities, there is an increasing need for a spatiotemporally detailed pipeline to analyze functional data for accurately characterizing of mTBI assessment.Objective: To develop and apply a personalized, multimodal analytic pipeline that integrates neurosource-driven EEG and MEG data to investigate functional brain dynamics, with the aim of supporting diagnosis and monitoring recovery in individuals with mTBI during the subacute phase (2–4 weeks post-injury).Methods: After a detailed preprocessing pipeline, three-minute segments of artifact-free resting-state EEG and MEG data were analyzed from individuals with mTBI during the subacute recovery phase, as well as from orthopedic control (OC) participants. Source localization was conducted using anatomically realistic, subject-specific head models generated through the combination of a new segmentation pipeline and the Finite Element Method (FEM). The realistically head volume conductor was then serve the base of the estimation of the neural activity using a Linearly Constrained Minimum Variance (LCMV) beamformer to improve spatial accuracy. Through the new segmentation pipeline, the cortex was divided into 94 regions of interest (ROI) selected from the AAL3 atlas. For each ROI, representative time series were extracted from the source-level signals and filtered into delta, theta, and alpha frequency rhythms. Dynamic Functional Connectivity Graphs (DFCGs) was estimated via amplitude envelope correlations using the Hilbert transform. The DFCGs were then filtered using Orthogonal Minimal Spanning Trees (OMST) to reduce spurious node connections. In the final step, the degree of DFCGs were converted to symbolic time series via the Neural-Gas (NG) algorithm and the Complexity Index (CI) was computed to quantify the richness and variability of neural dynamics across delta, theta, and alpha frequency bands.Results: The CI in the theta band (4–8 Hz) consistently reflected functional brain alterations and recovery following mTBI. Reductions in CI were observed across both EEG and MEG modalities, highlighting theta-band CI as a potential biomarker for early diagnosis and longitudinal monitoring of mTBI recovery.Novelty: This study introduces a multimodal, source-level pipeline that combines advanced head modeling with dynamic connectivity metrics to sensitively track mTBI-related brain alterations during the subacute phase, offering novel insights into neural recovery and identifying potential biomarkers for diagnosis and prognosis.

Διαθέσιμα αρχεία

Υπηρεσίες