I am a clinical epileptologist and physician scientist and spend the majority of my time doing neural engineering research. My dual training represents over 20 years of dedication to my career goal: to develop improved epilepsy devices and therapies with engineering tools and quantitative analysis of brain signals. My research integrates the wide range of training I have undertaken: clinical epileptology, computational neuroscience, advanced mathematical and engineering tools, basic electrophysiology, and translational research. From the beginning of my career, I have investigated the physiological methods by which neural ensembles detect and synchronize to rhythmic signals, and stimulation paradigms that can stop seizures.

Much of my work has focused on biomarkers of epilepsy such as High Frequency Oscillations (HFOs). This work spans animal models, computational models, and analysis of human data. Overarching all of these is developing quantitative tools that can describe the phenomena on analogous terms across all models and in humans. I have developed tools to automatically detect and process HFOs in human EEG. This combination of clinical, computational, and machine learning tools is crucial in understanding the mechanisms and features of epilepsy.

A separate line of research involves understanding and manipulating the underlying dynamics of seizures. My collaborators and I developed a novel method of characterizing seizures based upon their dynamics. Using seizure data from many species and human EEG, we modeled the onset and offset of seizures by focusing on the invariant properties of the most common bifurcations, then made a taxonomy of seizures in human epilepsy. The result was the Taxonomy of Seizure Dynamics, which found all predicted bifurcations in human seizures. This work was able to explain several unusual epileptic phenomena and has enabled a novel branch of epilepsy research in which we characterize how the brain state can move and influence seizure activity. Our ongoing collaboration, combining the disciplines of physics, engineering, neuroscience, and clinical epilepsy, is focused on leveraging this model to improve our understanding and treatment options in epilepsy.

Overlying all of our work is the goal of acquiring better neural data, interpreting it with advanced tools, and manipulating the system to control seizures better. This goal has led us to several novel quantitative tools and robust findings, each with the overriding goal of implementation into humans.

The lab uses a combination of electrophysiology, machine learning, signal processing, and computational modeling to model and describe neural data. Data for these projects are acquired from a large database of human patients, an ongoing clinical study in patients undergoing surgical implantation of electrodes, and several outside collaborations in other models. The lab is specifically researching the relationship of high frequency oscillations with seizure mechanisms, developing methods to target and stimulate the brain to stop seizures, and methods to quantify seizure dynamics.