Simplified-Delay-Multiply-and-Sum-Based Promising Beamformer for Real-Time Photoacoustic Imaging

Photoacoustic imaging (PAI) using a traditional delay-and-sum (DAS) beamformer suffers from various limitations such as low contrast, low signal-to-noise ratio (SNR), and strong sidelobe influence. Consequently, reconstructed image quality dramatically degrades. The present work improves reconstructed image quality by the use of a novel adaptive weighting algorithm named simplified-delay-multiply-and-standard-deviation (SDMASD). SDMASD combined with established DAS and delay-multiply-and-sum (DMAS) methodologies are presented in this article. This present method can efficiently improve the achievable resolution, contrast, and SNR and conquer all the persisting drawbacks present in the DAS and DMAS beamformers. Computational complexity in this method is reduced from
$O(N^{2})$
to
$O(N)$
using a unique mathematical tool. Therefore, it can be programmable for real-time PA image reconstruction. The performance of the algorithm has been accelerated using graphics processing unit (GPU) programming. It executes 50 times faster than the standard CPU-based DAS method. A validation study of the algorithm was carried out both numerically and experimentally. In the experimental study, a low-cost (16-element) linear array transducer in a home-built PAI system was used to acquire the photoacoustic (PA) signals. All the results illustrate an excellent capability of the proposed algorithm in improving image quality metrics. The quantitative evaluation of the agar phantom shows that SDMASD leads to an improvement of
${\sim } 45\%$
and
${\sim } 40\%$
in SNR, and
${\sim } 20\%$
and
${\sim } 30\%$
in full-width half maximum (FWHM) compared with DAS and DMAS, respectively. Conclusively, the study demonstrates that the proposed algorithm is very much powerful and promising for a cost-effective PAI system using a low-cost linear array transducer.