Abstract
Non-invasive photoacoustic imaging has the advantages of good ultrasonic resolution and high optical absorption contrast when compared with other blood-oxygen-saturation measurement techniques. The purpose of this study is to develop a photoacoustic-imaging-based quantitative measurement technique for the determination of blood oxygen saturation and verify its feasibility by computer simulation and phantom experiments. To perform the robust measurement, an optimized wavelengths set of exciting laser was chosen to improve the signal-to -noise ratio of photoacoustic signals and the stability of matrix inversion based on least-square method which was used to calculate blood oxygen saturation first. According to the penetration depth of photons at each wavelength in the set, a proper bandwidth of ultrasonic transducer was selected to retain the proportionality between the optical absorption coefficient of the interrogating blood and its corresponding photoacoustic signal amplitude. Furthermore, we proposed a mathematical model to get fluence-compensation coefficients with optimization algorithm to compensate its changes depending on different wavelengths. Simulation results indicated the saturation effect, which describes the violation of the linearity between the measured photoacoustic signal amplitude and the target object’s optical absorption coefficient, would happen with optical absorption coefficient increasing in target tissue. We also confirmed that transducers with larger bandwidth could provide more accurate estimation for blood oxygenation saturation when using the selected wavelength set. In the phantom experiments, two kinds of ink with distinct absorption spectra were used to mimic oxy-hemoglobin and deoxy-hemoglobin. The experimental results showed that transducers with larger bandwidth offered better estimation for the mixing ratio between the two kinds of ink with the selected wavelength set, which agreed with the simulation result and experimental result of ex vivo blood data. It demonstrated that the proposed photoacoustic measurement technique is capable of quantitative blood-oxygen-saturation measurement in single blood vessel. In addition, the differences of fluence at the interrogating blood under the selected wavelength set were simulated by Monte Carlo method. The distorted photoacoustic signal amplitude affected by differences of fluence was substituted into our mathematical model with optimization algorithm to get fluence-compensation coefficients. The result showed that our mathematical model with fluence-compensation coefficients provides a better estimation for blood oxygenation saturation compared with the original method. It demonstrated that our mathematical model has the potential to do quantitative blood-oxygen-saturation measurement for in vivo application.