The bandwidth limitation and aperture size of the transducer limits the resolution of a photoacoustic computed tomography system. If the separation between two sources is smaller than the point spread function width of the imaging system, they will appear as a single object at different wavelengths. It was shown previously in ultrasound motion imaging that phase difference between two consecutive frames can be used to detect lateral or axial motion with submicron resolution. We tested this method in the context of static PA imaging of two unresolved PA sources. We set up an experiment where we imaged a green and a yellow wire of 40 μm width with known relative absorption coefficients, separated by 355μm. Imaging was performed at 650nm and 460nm. The PA signal is recorded by a single element flat 1MHz transducer (Panametrics 0.5’’ V303) in the plane of the wires, so the targets are axially spaced seen from the transducer. We reconstructed the signals originating from both unresolved sources and measured the separation between them to be 350 µm. Similar performance was obtained using an array transducer, viewing the wires from the top so they were laterally separated in the imaging plane. The signal at two different wavelengths was recorded using a commercial imaging system. The two-wavelength phase difference at every pair of channels provides an estimate of the distance between the two absorbers, determined to be 350 µm by the median of the two-channel estimates.
|