Despite its enormous incidence, mild traumatic brain injury is not well understood. One aspect that needs more definition is how the mechanical energy during injury affects neural circuit function. Recent developments in cellular imaging probes provide an opportunity to assess the dynamic state of neural networks with single-cell resolution. In this dissertation, we developed imaging methods to assess the state of dissociated cortical networks exposed to mild injury. We probed the microarchitecture of an injured cortical circuit subject to two different injury levels, mild stretch (10% peak) and mild/moderate (35%). We found that mild injury produced a transient increase in calcium activity that dissipated within 1 h after injury. Alternatively, mild/moderate mechanical injury produced immediate disruption in network synchrony, loss in excitatory tone, and increased modular topology, suggesting a threshold for repair and degradation. The more significant changes in network behavior at moderate stretch are influenced by NMDA receptor activation and subsequent proteolytic changes in the neuronal populations. With the ability to analyze individual neurons in a circuit before and after injury, we identified several biomarkers that confer increased risk or protection from mechanical injury. We found that pre-injury connectivity and NMDA receptor subtype composition (NR2A and NR2B content) are important predictors of node loss and remodeling. Mechanistically, stretch injury caused a reduction in voltage-dependent Mg²⁺ block of the NR2B-cotaning NMDA receptors, resulting in increased uncorrelated activity both at the single channel and network level. The reduced coincidence detection of the NMDA receptor and overactivation of these receptors further impaired network function and plasticity. Given the demonstrated link between NR2B-NMDARs and mitochondrial dysfunction, we discovered that neuronal de-integration from the network is mediated through mitochondrial signaling. Finally, we bridged these network level studies with an investigation of changes in neurobehavior following blast-induced traumatic brain injury (bTBI), a form of mild TBI. We first developed and validated an open-source toolbox for automating the scoring of several common behavior tasks to study the deficits that occur following bTBI. We then specifically evaluated the role of neuronal transcription factor Elk-1 in mediating deficits following blast by exposing Elk-1 knockout mouse to equivalent blast pressure loading. Our systems-level behavior analysis showed that bTBI creates a complex change in behavior, with an increase in anxiety and loss of habituation in object recognition. Moreover, we found these behavioral deficits were eliminated in Elk-1 knockout animals exposed to blast loading. Together, we merged information from different perspectives (in silico, in vitro, and in vivo) and length scales (single channels, single-cells, networks, and animals) to study the impact of mild traumatic brain injury on neuronal networks and neurobehavior.