A low-rank-based image reconstruction method for dynamic contrast-enhanced multispectral optoacoustic tomography (DCE-MSOT)

This is the authors' abstract. We don't add key points for this paper.

Dynamic contrast-enhanced multispectral optoacoustic tomography (DCE-MSOT) is a non-invasive imaging modality that uses optical excitation along with acoustic detection to estimate tumor perfusion rates in biological tissues. Image formation is based on the optoacoustic effect and uses measurements gathered at multiple views and incident wavelengths. The reconstructed image can then be viewed as a function of space, wavelength, and time. Reconstructing such multidimensional images presents significant computational and memory burden. To develop efficient accurate reconstruction methods, it is mandatory to exploit the underlying redundancies in the image due to correlations across the spatial, temporal, and spectral dimensions. Previous works have demonstrated that a low-rank-based representation of the sought-after spatiotemporal object can achieve accurate, memory-efficient reconstruction of high-resolution dynamic images. This work further extends low-rank approaches to multispectral dynamic image reconstruction. The proposed methods are demonstrated using in-vivo imaging studies of tumor perfusion in small animal models using DCE-MSOT. Comparisons with conventional frame-by-frame image reconstruction show that the proposed methods are more accurate (higher goodness of fit) at estimating the normalized tumor perfusion wash-in and wash-out rates, while requiring lesser computational resources.