Quantitative photoacoustic image reconstruction improves accuracy in deep tissue structures.

Michael A Mastanduno,Sanjiv S Gambhir

doi:10.1364/boe.7.003811

Abstract

Photoacoustic imaging (PAI) is emerging as a potentially powerful imaging tool with multiple applications. Image reconstruction for PAI has been relatively limited because of limited or no modeling of light delivery to deep tissues. This work demonstrates a numerical approach to quantitative photoacoustic image reconstruction that minimizes depth and spectrally derived artifacts. We present the first time-domain quantitative photoacoustic image reconstruction algorithm that models optical sources through acoustic data to create quantitative images of absorption coefficients. We demonstrate quantitative accuracy of less than 5% error in large 3 cm diameter 2D geometries with multiple targets and within 22% error in the largest size quantitative photoacoustic studies to date (6cm diameter). We extend the algorithm to spectral data, reconstructing 6 varying chromophores to within 17% of the true values. This quantitiative PA tomography method was able to improve considerably on filtered-back projection from the standpoint of image quality, absolute, and relative quantification in all our simulation geometries. We characterize the effects of time step size, initial guess, and source configuration on final accuracy. This work could help to generate accurate quantitative images from both endogenous absorbers and exogenous photoacoustic dyes in both preclinical and clinical work, thereby increasing the information content obtained especially from deep-tissue photoacoustic imaging studies.

Highlights

Photoacoustic imaging (PAI) is an exciting, emerging imaging modality that has the ability to combine the high intrinsic contrast of optical imaging with the excellent spatial resolution of ultrasound imaging [1,2,3,4]
Three experiments were completed to characterize the effect of initial guess and time resolution on the completed solution. 3.1 Three target reconstruction The reconstruction shown in Fig. 1 demonstrates the Quantitiative Photoacoustic Tomography (qPAT) algorithm’s ability to quantitatively reconstruct absorbing targets at multiple depths using pressure data generated from optical sources
This work has demonstrated a numerical approach to quantitative photoacoustic image reconstruction that minimizes depth and spectrally derived artifacts when compared with filtered back projection

Summary

Introduction

Photoacoustic imaging (PAI) is an exciting, emerging imaging modality that has the ability to combine the high intrinsic contrast of optical imaging with the excellent spatial resolution of ultrasound imaging [1,2,3,4]. PAI and Photoacoustic Tomography (PAT) have been developing for many years in both clinical and preclinical regimes with applications in skin [6] and breast cancer [7,8], lymph node mapping [9,10,11], and endoscopy [12,13]. These technique relies on absorption of light to create contrast, a phenomenon that is governed by both tissue optical properties and light fluence. Quantitative images of absorption and spectrally derived chromophores at unprecedented depths

Objectives

Methods

Results

Discussion

Conclusion