Improved Synthetic Aperture Focusing Technique for Acoustic-resolution Photoacoustic Microscopy

Acoustic-resolution photoacoustic microscopy (AR-PAM) capable of imaging micro-vessels with high resolution and high contrast has shown its potential in functional neuro-vascular imaging, dermatology and related cancer research. In AR-PAM, a mechanically scanned, single crystal transducer with a fixed focus is commonly used, which causes the image quality to deteriorate significantly in the out-of-focus region. A virtual-detector based synthetic aperture focusing technique (VD-SAFT) was used previously to extend the depth of focus of the AR-PAM where the focal point of the transducer is treated as a virtual point detector. However, the performance of VD-SAFT is impaired by its high sidelobes. Such high sidelobes result from over-simplified virtual-point-detector approximation of the focal point. To solve this problem, we introduce an improved synthetic aperture focusing technique (ISAFT) for AR-PAM, which considers diffraction effect in the vicinity of the VD, i.e., the focal point. This technique is based on a linear, discrete model of the AR-PAM system developed using matrix formalism. Using this model, a spatiotemporal optimal filter in minimum mean square error sense is designed to deconvolve the spatial impulse responses associated with the focused transducer at every imaging point; thus retrospective focusing can be achieved. The performance of the proposed ISAFT is verified using simulation data. A 25-MHz AR-PAM with a single element transducer which has a 6-mm diameter and a 12-mm focal depth (i.e., f-number = 2) is simulated. The improvement in spatial resolution and sidelobe suppression of ISAFT is superior to that of VD-SAFT because the full geometry of the VD, instead of the simplified virtual-point-detector approximation, is taken into consideration. It is demonstrated by simulation that the proposed method effectively improves the degraded lateral resolution in the out-of-focused region for AR-PAM. Future work will focus on the phantom and in vivo experimental verifications and the study of the effect of signal-to-noise ratio on the ISAFT.