What is the Neuro-intergrative Connectivity Platform (NIC)?

Updates

Overview

The NIC Platform is a comprehensive user-friendly platform to analyze functional connectivity in epilepsy patients. The platform converts EDF files, calculates correlation measures, and allows management of past workflow runs.

Epilepsy

Epilepsy affects nearly 50 million individuals worldwide. Demographically, it affects all age groups and genders. While certain patients are able to manage their condition with a single medication, many need multiple therapies including neuromodulation devices, surgery, and dietary restrictions. Even with current advances, 1/3 of patients still have uncontrollable seizures. Research into seizure pathophysiology have involved analysis of genetic, electrophysiological, and mechanistic components.

Functional Connectivity

Functional connectivity is a statistical measure that captures correlation between different, often spatially separate, regions in the brain. Often it is measured by correlation or coherence. Functional connectivity can be used to determine signal propogation in patients with epilepsy, hopefully leading to understanding the underlying mechanisms of the disease.

Features of the NIC Workflow

Conversion

EEG files are generally disseminated in a binary format known as European Data Format (EDF). Due to a loss of interpretability, the Sahoo Lab has developed a JSON-based human-readable format we call Cloudwave Signal Format. The NIC workflow allows for conversion from EDF to CSF files using an intuitive user interface.

Correlation

To investigate functional connectivity in the brain, statistical correlation measures are used. The NIC workflow supports the following measures: Pearson Correlation, Mean Phase Coherence, and Non-linear correlation by Pijn et al.

The output is returned to the user for further evaluation.

Algebraic Topology

To analyze the change in network graph topology as the seizure progresses, we use methods in algebraic topology. Namely, the NIC workflow calculates the Betti numbers given temporal input parameters.

Provenance

Finally, due to the rampant lack of scientific reproducibility in biomedicine generally, we have developed the NIC platform to ensure studies and analysis can be recreated. We do this through the use of Provenance, such as patient metainformation and past computation run history.