I-Corps: Translation potential of a computational platform that simulates how cortical neural circuits generate electroencephalogram (EEG) signals

Award ID: 2627393

This I-Corps project is based on the development of a brain simulation platform that models brain circuits to simulate measurable characteristics associated with brain illness and the effects of therapeutic compounds. Identifying the right therapy for the right patient is one of the biggest challenges facing drug developers and clinicians. This technology aims to provide treatment response predictions that will aid diagnostics and clinical psychiatry and neurology research and practice. It is designed to integrate realistic neural modeling with artificial intelligence to support mechanistic interpretation of brain signals to strengthen the evidence base for drug development and improve the consistency of clinical decision-making. This system may help clinicians and researchers identify more effective therapies for individuals living with psychiatric and neurological disorders by generating insight into how different neural circuits contribute to observed patterns of electrical activity of the brain. This I-Corps project utilizes experiential learning coupled with first-hand investigation of the industry ecosystem to assess the translation potential for using electroencephalography in clinical decision support by combining high fidelity neural modeling with state of the art machine learning. This is a physics-based computational platform that simulates how cortical neural circuits generate electroencephalogram (EEG) and magnetoencephalography (MEG) signals. Unlike black-box machine learning approaches that identify correlations without mechanistic insight, this technology models biophysically realistic pyramidal neurons and interneurons organized in cortical layers, enabling researchers to directly link cellular and synaptic changes to measurable brain signals. Combining this biophysical modeling framework with machine learning methods allows for biomarker identification and model specification as well as mechanistic interpretability of drug effects on neural circuits, cross-species translation from rodent to human, and personalized parameter estimation from individual patient recordings. This approach may enable prediction of treatment response and support development of neurotherapeutics that are better matched to disease mechanisms. In addition, this may lead to improved clinical precision in psychiatry and neurology, more efficient therapeutic development, and enhanced computational tools for interpreting complex brain signals. These advances may support more personalized care and more effective translation of neuroscience research into clinical practice. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

NSF Program Director: Ruth Shuman