Feasibility of quantitative tissue characterization using novel parameters extracted from photoacoustic power spectrum

Abstract

Photoacoustic (PA) spectral parameters, obtained using conventional frequency domain analysis of photoacoustic power spectrum, bear complex relationship with absorber size, due to which it is difficult to recover quantitative information related to tissue microstructure accurately, using the conventional spectral parameters. In this study, we proposed to utilise two new spectral parameters namely number of lobes as well as average lobe width, which may change linearly with PA absorber radius, for recovering quantitative values of PA absorber size. Using an analytical model of PA power spectrum with bandlimited acquisition, we developed linear relationships between number of lobes and absorber radius as well as average lobe width and absorber radius for a wide range of radius values. These linear relationships were validated using simulations and experimentally acquired PA signal. The preliminary results of the simulation as well as phantom study showed that using the computed values of number of lobes as well as average lobe width in the linear relationships, absorber radius could be recovered with reasonable degree of accuracy. The proposed parameters were extracted from the multiwavelength PA data generated by freshly excised prostate tissue specimens from 30 human patients undergoing prostatectomy for biopsy confirmed prostate cancer. Statistical analysis showed that both the proposed parameters were significantly different between malignant and non malignant prostate as well as malignant and normal prostate. The preliminary findings showed that the proposed parameters can be used to differentiate diseased tissue pathology form normal tissue.