Three-dimensional photoacoustic imaging by sparse-array detection and iterative image reconstruction

Abstract

Photoacoustic imaging (PAI) has the potential to acquire 3-D optical images at high speed. Attempts at 3-D photoacoustic imaging have used a dense 2-D array of ultrasound detectors or have densely scanned a single detector on a 2-D surface. The former approach is costly and complicated to realize, while the latter is inherently slow. We present a different approach based on a sparse 2-D array of detector elements and an iterative reconstruction algorithm. This approach has the potential for fast image acquisition, since no mechanical scanning is required, and for simple and compact construction due to the smaller number of detector elements. We obtained spatial sensitivity maps of the sparse array and used them to optimize the image reconstruction algorithm. We then validated the method on phantoms containing 3-D distributions of optically absorbing point sources. Reconstruction of the point sources from the time-domain signals resulted in images with good contrast and accurate localization (< or =1 mm error). Image acquisition time was 1 s. The results suggest that 3-D PAI with a sparse array of detector elements is a viable approach. Furthermore, the rapid acquisition speed indicates the possibility of high frame rate 3-D PAI.

Compressed sensing in photoacoustic tomography in vivo

“…For example, for a fast 512-element ring array PAT system with a 64-channel data acquisition module, 22 it takes 8 laser shots to collect data from all 512 elements. For direct 3-D reconstruction PAT applications, 23,24 the data from a 2-D ultrasonic array is usually an extremely sparse sampling of the detecting aperture. Moreover, channel cross talk is also related to the space between neighboring elements ͑kerf͒, and an extensive spatial sampling may increase the cross talk.…”

In vivo optical-resolution photoacoustic computed tomography with compressed sensing

“…In addition, densely packed high-frequency ultrasonic arrays (with a pitch ∼1=2 of the ultrasonic center wavelength), needed for obtaining high-quality OR-PACT images with backprojection (BP) reconstruction, are still difficult to fabricate and can be quite expensive. While the use of a sparse ultrasonic array for PACT has also been explored, the image quality was somewhat compromised with conventional BP reconstruction [4].…”

Numerical investigation of the effects of shear waves in transcranial photoacoustic tomography with a planar geometry

“…The utility of PAT has been demonstrated in a number of in vivo studies of biological structure and function [5][6][7] and in facilitating a number of medically relevant diagnostic tasks. 2,[8][9][10][11][12][13] Functional imaging of mouse brains has been successfully demonstrated with PAT. 5,6 For larger animals with thicker skulls, PAT has been demonstrated to provide qualitatively useful images of brain structures; [14][15][16][17] however, the transmission of ultrasonic waves through the skull bone(s) induces strong changes to the photoacoustic signal through the processes of absorption, strong reflections, and longitudinal-to-shear mode conversion.…”