Gated DOT

Abstract: Diffuse optical tomography (DOT) is a promising technique for through skull high-resolution neural imaging. Recent work known as time-of-flight diffuse optical tomography (ToF-DOT) has demonstrated that the addition of the temporal/transient/time-of-travel dimension can significantly improve the imaging performance of DOT through a skull phantom. Unfortunately, the addition of this transient dimension comes with additional challenges with regards to (a) significantly increased hardware complexity, (b) constraints on the maximum number of source-detector pairs that can be utilized, and (c) requirements for significant memory and computation. Together these challenges limit the performance improvements that can be achieved in practice. Motivated by the promise of ToF-DOT, we introduce Gated DOT (or GDOT). Instead of measuring photon arrival times to estimate the entire transient response, we propose to measure the total photon intensity within a specific time-of-flight window, i.e. a time gate. This ensures that only a single measurement per source-detector pair is collected. This simple modification (a) significantly reduces the hardware complexity of the sensor system, (b) allows for dramatic order of magnitude increases in source-detector pairs that can be used, and (c) achieves order of magnitude reductions in computational complexity. The increased number of source-detector pairs improves high resolution image particularly in moderately-deep imaging regimes like through skull imaging (5-10 mm). We demonstrate both via simulation and through a benchtop experimental prototype system the many performance advantages of GDOT over both DOT and ToF-DOT in the application space of neural imaging.

[under review]