Spatial and temporal modulation in time-domain diffuse optical tomography

Diffuse optical tomography (DOT) utilizes visible and near-infrared light to non-invasively study biological structures and functions, such as imaging breast cancer and brain activity. Time-domain DOT systems use pulsed light and measure the time-varying temporal point spread function. Time-domain DOT provides information from both superficial and deep tissue, and thus it is regarded as a technique that provides the richest information, among other measurement types in DOT, such as continuous-wave and intensity-modulated DOT. However, modeling and image reconstruction of time-domain data as such is computationally expensive. To overcome this problem, different transformations of data types in DOT have been utilized with an aim to compress the data while maintaining the rich information content. Furthermore, different structural illuminations have been utilized in DOT to enhance the imaging modality. In this work, we study the use of temporal and spatial modulation in time-domain DOT using numerical simulations. For temporal modulation, a truncated Fourier-series approximation is utilized and the use of different numbers of frequencies is studied. Furthermore, structural illumination and detection are implemented using Hadamard patterns. The use of different numbers and different complexities of patterns is studied. The simulations show that the resolution and contrast of reconstructed absorption and scattering images can be improved by increasing the number of Fourier frequencies and the number of Hadamard patterns in the data. However, it was also noticed that, after a certain level of improvement, including more frequencies and Hadamard patterns do not change reconstructed images significantly.