Limited data image reconstruction in optoacoustic tomography by constrained, total variation minimization

Optoacoustic Tomography (OAT) is an emerging hybrid imaging technique with great potential for a wide range of biomedical imaging applications. Assuming point-like transducers, analytic algorithms are available for image reconstruction, but they are applicable only when the measured data are densely sampled on an aperture that encloses the object. In many cases of practical interest, however, measurements may be limited in number and are acquired on an incomplete aperture. Total variation (TV) minimization has been proved to be a powerful tool for limited-data reconstruction. However, most previous studies of limited-data OAT were based on an approximate imaging model that assumed point-like transducers, which limits the improvements on the reconstructed OAT image quality. In this work, we develop and investigate an iterative reconstruction algorithm incorporating ultrasonic transducer properties applicable for limited-data OAT. The algorithm is based on the minimization of the image TV subject to a data consistency condition, and is conceptually and mathematically distinct from classic iterative reconstruction algorithms. Preliminary computer-simulation studies are conducted to investigate the proposed algorithm. These studies reveal that the constrained, total variation minimization algorithm can yield accurate reconstructions in many limited-data applications where classic algorithms do not perform well.