Abstracts

Rationale: While changes in diffuse cortical reflectance have been well described in both animal and human seizures, the individual contributions of absorption and scattering have not hitherto been explored. We introduce novel imaging platforms using spatial modulation of near-infrared light (NIR) to separate and quantitatively map absorption and scattering changes before and during seizure activity in two distinct seizure models.Methods: Generalized seizures were induced in mice using pentylenetetrazol (PTZ) (100 mg/kg, IP). Continuous electroencephalographic (EEG) recordings were obtained through anterior and posterior cortical tungsten microelectrodes connected to a digital EEG acquisition system. Focal cortical seizures were induced in a separate group of mice using cortical microinjection of 4-aminopyridine (4-AP) (150 nL, 25 mmol in normal saline) in the region of the anterior electrode. Seizures were reversed with pentobarbital (PB) (30 mg/kg, IP). Ictal onset (SZ) was determined post hoc by an observer blinded to the optical data and by power analysis of EEG epochs. Spatially modulated imaging (MI), a novel optical technique developed at the Beckman Laser Institute at UC Irvine, uses projected patterns of light and dark bands to enable quantitative maps of optical scattering and absorption for the first time. Images were obtained through cranial windows every 5.3-9.7 seconds at 750 and 850 nm using a spatial frequency of 0.34 mm-1 in the PTZ-injected mice. In the 4-AP model, the MI system was coupled to a computed tomography imaging spectrometer (CTIS) to allow simultaneous spectral acquisition of multiple wavelengths in the NIR range with faster acquisition times (1-2 seconds). Post-processing software (Matlab) was used to generate maps and time plots of scattering coefficients and chromophore concentrations (e.g. total hemoglobin). Results: PTZ injection reliably produced generalized seizures. Cortical 4-AP microinjection induced propagating cortical seizures, verified through delayed electrographic onset in posterior electrodes distant from the injection site. In each model (n=6), a marked decrease in the scattering coefficient preceded electrographic seizure onset by tens of seconds, with a further precipitous decrease following seizure onset(Figure 1:MI/PTZ; Figure2:MI-CTIS/4-AP). Scattering changes were restored to baseline following seizure termination.Conclusions: Previously, the individual contributions of light absorption and scattering have been impossible to distinguish with methods such as intrinsic optical signal (IOS) imaging. Using the entirely novel methods of MI and CTIS, we describe the individual contribution of light scattering to the optical signal change before and during seizure activity. Preictal reduction in optical scattering was observed in both generalized (PTZ) and focal (4-AP) seizure models. These results coincide well with previous findings of preictal constriction in brain extracellular space, and provide proof of principle for optical detection of the pre-seizure state on a clinically relevant timescale.