Introducing Cerebrum

submited by

Style Pass

2024-12-24 18:30:05

Advancements in computational neuroscience are continually reshaping our understanding of the brain’s intricate networks. A key challenge in this field is deciphering the dynamic connectivity of neural networks, which is essential for both fundamental neuroscience and the development of clinical applications. To address this, we introduce Cerebrum, a novel framework that combines biologically inspired neuron models with cutting-edge machine learning techniques to simulate and infer synaptic connectivity in large-scale brain networks.

Traditional approaches to studying brain networks often rely on graph theoretical methods that provide valuable insights into the static topological properties of neural connections. However, these methods typically overlook the temporal dynamics that are crucial for understanding how neuronal activity evolves over time. Cerebrum bridges this gap by integrating the Hodgkin-Huxley (HH) neuron model, known for its biological realism, with Graph Neural Networks (GNNs), which excel at learning complex patterns in graph-structured data. The HH model simulates the electrical characteristics of neurons, capturing the essential dynamics of action potentials and synaptic interactions. By combining this with GNNs, Cerebrum can effectively learn from simulated neuronal activity to predict the underlying synaptic connectivity. This integration allows for a more comprehensive analysis of brain networks, considering both their structural and dynamic properties.

To train and evaluate Cerebrum, we employed three canonical network topologies: Erdős-Rényi, Small-World, and Scale-Free. These topologies serve as ground-truth connectivity matrices, providing diverse structural frameworks for the GNNs to learn from. Each topology presents unique characteristics. Erdős-Rényi networks consist of randomly connected nodes, offering a baseline for connectivity without inherent clustering or hierarchy. Small-World networks feature high clustering and short path lengths, mimicking the efficient communication pathways observed in many biological networks. Scale-Free networks are characterized by the presence of hub nodes with a high degree of connectivity, reflecting the hierarchical organization seen in various neural systems. By training Cerebrum on these different topologies, we can assess its ability to generalize across diverse network structures and better understand the factors that influence connectivity inference.