Optimizing Memory via Real-Time Monitoring of Neural Data with Adaptive Brain Stimulation Machine Learning Algorithms

Overview

In this research, I aim to test how stimulation of distinct brain systems affects memory organization and whether such stimulation exhibits state -dependent effects. Aging and neurological disorders lead to neurodegeneration that decreases our ability to re member events. Deep brain stimulation (DBS) has emerged as a potential solution by stimulating neurons with electricity to enhance neuronal activity during memory tasks. I hypothesize that DBS can improve memory when targeted towards poor memory encoding states by influencing semantic organization. This study utilized data from 38 neurosurgical epilepsy patients with implanted electrodes performing the free-recall tasks with and without stimulation. Spectral data from each electrode were used in a user-specific algorithm based on the random forest to predict the memory encoding state at an average area under the curve (AUC) of 98% across all subjects. Next, a DBS success classifier based on the random forest predicted improvements in memory outcomes when stimulation was applied based on pre-stimulus memory encoding state qualities found through the user -specific algorithms, achieving an average of 84% AUC. This study established that stimulation targeted at memory networks enhances memory by targeting poor encoding states. In addition to state dependency, it was discovered that stimulation applied to the left lateral temporal cortex (LTC) enhanced the brain’s ability to create semantic clusters and recall performance. The combination of these two algorithms with the identification of the positive effects on the left LTC indicates a promising relationship between targeted DBS and memory enhancement, proposing a new treatment for improved episodic verbal memory.

Competition history

• JSHS 2024 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.