University of Twente Student Theses
Relatie tussen MRI-features en EEG-patronen bij postanoxische comapatiënten
Kortstee, Lauren and Melchers, Jens and Boeije, Lisanne and Rooij, Lieke de (2025) Relatie tussen MRI-features en EEG-patronen bij postanoxische comapatiënten.
Full text not available from this repository.
| Full Text Status: | Access to this publication is restricted |
| Abstract: | Background: After successful resuscitation from out-of-hospital cardiac arrest, many patients remain comatose due to severe post-anoxic encephalopathy. Only about 50% of these patients recover, while the rest either die or remain permanently dependent on care. EEG is a key tool in prognostication, often complemented by neurological assessment. Prior studies suggest that MRI can offer insights into the structural and functional brain changes underlying EEG patterns. This study investigates which MRI features correlate with prognostic EEG patterns in post-anoxic comatose patients. Methods: This retrospective study included patients from Rijnstate Hospital (Arnhem) and Radboudumc (Nijmegen) who underwent continuous EEG monitoring during the first three days post-cardiac arrest. MRI scans were acquired between 1.5 and 10 days after the event. From these, 27 clinically relevant features were extracted. Random forest models were trained on all possible combinations of three features to classify patients into four EEG-defined groups. The models were validated using stratified 5-fold cross-validation, and SHAP values were used to assess feature importance. Results: Among 76 patients, the combination of three MRI features—default mode network intensity in the posterior cingulate cortex, mean diffusivity in the occipital cortex, and fractional anisotropy in white matter—achieved the highest discrimination (macro-AUC: 0.83; 95% CI: 0.73–0.95). Conclusion: This MRI feature combination most effectively distinguishes between prognostic EEG patterns in post-anoxic coma. |
| Item Type: | Student Thesis (Bachelor) |
| Clients: | Rijnstate, Arnhem, Nederland |
| Faculty: | TNW: Science and Technology |
| Subject: | 44 medicine, 50 technical science in general |
| Programme: | Technical Medicine BSc (50033) |
| Link to this item: | https://purl.utwente.nl/essays/106654 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page