Projects

An exploratory multimodal machine learning approach examining potential relationships between brain features and cognitive training outcomes in stroke patients, integrating structural MRI, blood biomarkers, and normative deviation metrics.

View full project

Project at Simula Summer School in Computational Physiology (SSCP) investigating how artificial neural networks can model hippocampal memory functions. Implementing and experimenting with Hopfield Networks and Kanerva Machines to simulate pattern completion, pattern separation, and associative memory processes that underlie spatial and episodic memory formation.

View full project

UiO:Life Science summer research project analyzing stereoelectroencephalography (SEEG) data to investigate hippocampal memory circuits and their input regions. Focusing on theta oscillations across the hippocampus and its primary input regions (entorhinal cortex and perirhinal cortex) during novel image processing, this project develops automated pipelines to extract, filter, and analyze neural activity patterns in the medial temporal lobe memory system.

View full project

Enhancing Kernel SHAP interpretability for neuroimaging by addressing feature independence assumptions through domain-specific constraints. Using Markov Random Fields with a graph-theoretical approach to incorporate spatial and connectivity awareness for more biologically plausible explanations.