Modeling the shape of cylindrically focused transducers in three-dimensional optoacoustic tomography

Cross sectional tomographic systems based on cylindrically focused transducers are widely used in optoacoustic (photoacoustic) imaging due to important advantages they provide such as high-cross sectional resolution, real-time imaging capacity, and high-throughput performance. Tomographic images in such systems are commonly obtained by means of two-dimensional (2-D) reconstruction procedures assuming point-like detectors, and volumetric (whole-body) imaging is performed by superimposing the cross sectional images for different positions along the scanning direction. Such reconstruction strategy generally leads to in-plane and out-of-plane artifacts as well as significant quantification errors. Herein, we introduce two equivalent full three-dimensional (3-D) models capable of accounting for the shape of cylindrically focused transducers. The performance of these models in 3-D reconstructions considering several scanning positions is analyzed in this work. Improvements of the results rendered with the introduced reconstruction procedure as compared with the 2-D-based approach are described and discussed for simulations and experiments with phantoms and biological tissues.