Abstracts

Rationale: Our previous study showed that photoacoustic tomography (PAT), an emerging imaging modality with unique ability of imaging biological tissues with high optical contrast and high ultrasound resolution, is able to image seizure-onset zone in an animal model of focal seizures. To provide three-dimensional (3D) real-time hemodynamic information such as blood oxygenation and blood volume, a PAT system based on a cylindrical ultrasound transducer array has been built. Tissue-like phantom has been used to validate the high speed imaging system and in vivo animal imaging has been performed using a bicuculline methiodide (BMI) animal model of epilepsy. Methods: A cylindrical transducer array with 192 elements was employed for collecting the light induced acoustic signals. The laser beam from a pulsed Nd:YAG laser with 4ns pulse duration was expanded to 3cm diameter thus to apply a light power of 15mJ/cm2 at the surface of the phantom/rat head, which was lower than the maximum permissible exposure for skin (20 mJ/cm2). The collected acoustic signal was amplified by 192-channel preamplifiers and then digitized by a 64-channel parallel data acquisition system with 50MHz sampling rate and additional 60dB gain. A full set of 3D photoacoustic data can be obtained at every three fires of the laser. Tissue-like phantoms have been used to validate and optimize this fast imaging system. Animal PAT data were obtained from several young male rats weighing 50~60g using urethane anesthesia (1mg/g). Acute seizure foci were induced by the injection of BMI over the frontal neocortex and confirmed with EEG. Continues photoacoustic data was been collecting for 30 minutes after the BMI injection. Results: The tissue-like phantom results obtained showed that the fast PAT system could obtain high spatial resolution absorption images at 10Hz. In the animal experiments, focal seizures were induced by microinjections of BMI into one side of parietal cortex and confirmed with EEG recordings which showed high amplitude spike and wave discharges. PAT images were recovered from data obtained at 10Hz within 30 minutes after the BMI injection. Large increase of absorption was observed in a region around the location of BMI injection in two to five minutes which reflected the increasing of blood flow in the same area. These seizures were accurately localized by PAT. The dynamic changes in PAT imaging were comparable to EEG changes during seizures. Conclusions: Our results showed that by using the fast PAT system seizure foci in rats could be imaged at a speed of 10Hz and a spatial resolution of 400?m. The imaging speed of the system can be improved to up to 200Hz by employing a faster laser source available commercially. The ability of functional imaging at a spatial resolution comparable to MRI and at a temporal resolution comparable to EEG makes high speed PAT a powerful tool in the studies of epilepsy.