Research On The Reconstruction Algorithm And Experiments Of Photoacoustic Imaging Technology

Based on photoacoustic effect, photoacoustic imaging (PAI) technology has been progressed fast in the field of bio-medical imaging in recent years. PAI is capable of mapping distribution of optical absorption within tissues, when irradiated by a pulsed laser. The technique uses the ultrasound signals as the carrier, which is excited by the transit absorption of light energy, to reconstruct the three dimensional PAI image, representative of the optical absorption distribution within the irradiated tissue. As such, PAI imaging technique combines the advantages of both optical and ultrasound imaging, providing images with high resolution and high contrast within deep tissue.This dissertation has a focus on the development of correlation theories for PAI imaging and the investigation of potential applications of PAI for bio-medical imaging. Based on the tissue optical properties, acoustic properties and the generation mechanism of photoacoustic signals, this thesis provides a theoretical analysis of the signal processing within both time and frequency domains, from which a modified reconstruction algorithm suitable for PAI has been proposed. By combining the numerical simulations on computer with the phantom experiments, the proposed modified reconstruction algorithm is tested and validated with considerations of some practical problems. In addition, this dissertation also deals with the improvement of PAI systems for 3D imaging of biological tissue, and the experimental approach that involves three kinds of biological tissues. All of the above make necessary basis of theories and experiments for further application study of PAI. The main tasks are summarized as below:1) It is found that the ill-condition of deconvolution problem in the system can be solved effectively by regularization method. Depending on numerical simulations, the regularization processing to the output signal is able to suppress the high frequency noise components and recover the input signal.2) The signal properties of multiple-bandwidth of round absorber with different diameters are investigated by numerical simulations and experimental approaches by use of appropriate phantoms. According to the corresponding diameter, the energy within three different frequency bands 50KHz~300KHz,300KHz~2MHz and 2MHz~6MHz, is normalized. This normalization results in a curve that describes the relationship between the energy ratio of photoacoustic signals in different frequency band and their corresponding diameters, from which the value of diameter that corresponds to the max energy ratio can be obtained. It is reported that the proposed normalization procedure is agreed well with the experimental results by use of appropriate phantoms.3) By the concept of algebraic reconstruction techniques (ART), the reconstruction algorithm with measurement in limited-views is investigated. The algorithm is designed to be applicable in PAI. The photoacoustic signals measured in limited-view are used to obtain the reconstructed image. The resolution of the image is analyzed to be 0.1mm. The reconstruction algorithm with measurement in limited-view has been applied to image eye, the result of which is satisfactory.4) By use of sound absorption material, the three-dimensional PAI with unfocused ultrasound transducer is proposed for the first time. This model of sound absorption not only confines the radiation direction of ultrasound wave of target subject, but also restricts the receipt angle of unfocused transducer to a small range. The tomographic performance of the model of sound absorption is analyzed by theoretical computation, followed by the 3D PAI experiments.5) The correction for acoustic speed in reconstruction algorithm for the medium with heterogeneous acoustic properties is implemented by a discretization method. By the correction for acoustic speed, the actual time-of-flight is obtained and the image refecting the real structure is reconstructed. This modification is validated by phantom experiments, while the imaging resolution is not compromised.6) PAI experiments for three kinds of biological tissues including chick embryo, porcine eye in-vitro and rat brain in-vivo are conducted with satisfactory results obtained.7) The optical clearing agent to be applied in PAI is proposed for the first time. The agent is proved to be effective by PAI experiment with porcine muscle. Based on this agent and PAI system, the imaging of rat brain function in-vivo is performed. And the reconstruction image reflects the cerebral blood vessels of rat well.