The overall goal of this project is to enable new methods in neurocritical care through the development of a quantitative framework for modeling human brain dynamics from electrophysiological data. This goal is timely because of an increasing emphasis in neurocritical care on tracking and predicting secondary injury and optimizing functional outcomes, needs that are challenging, unmet and consequential. Indeed, secondary neurological injuries and neurological worsening, including seizures, contribute to cognitive deficits and life-long impairments. The high prevalence of traumatic brain injuries means that the societal burden of these sequalae is significant, and their economic burden is measured in the tens of billions of dollars. Thus, technology that can help clinicians anticipate secondary injuries would provide tremendous leverage for improving the care of critically ill patients. Electrophysiological monitoring, such as the electroencephalogram (EEG), can provide a means to noninvasively examine brain electrical activity in high-risk populations, thus providing a pathway to such technology. However, many paradigms in EEG signal processing and informatics struggle with patient individuality and variation, compounded by the idiosyncrasy of neurological disease. As a result, many current approaches are limited to population-level characterizations and are susceptible to bias in datasets. In the proposed work, we will pursue a new approach to modeling human brain dynamics that is interpretable and robust to individual variability. We will develop and use dynamical systems models that generate simulated EEG activity, then fit these models to patient data in a minute-to-minute timescale. These models capture the internal biophysics of populations of neurons and are thus explanatory from a physiological perspective, setting the stage for innovative clinical applications. Specifically, we will: (i) develop the modeling framework and use it to quantify dynamical effects of brain injury in individual patients, (ii) track changes in internal brain dynamics and states through the course of injury, to predict neurological worsening, and (iii) use the model as an in silico surrogate to predict the response of patients to clinical interventions. We will validate our technical approach through retrospective and prospective studies in a heterogenous dataset spanning injuries and age. If successful, this work would represent a first-of-its-kind demonstration of a data-intensive approach to provide actionable clinical information about individual neurocritical care patients.