Evaluation Of Piezoelectric Micromachined Ultrasonic Transducer And Image Reconstruction Algorithm Of Photoacoustic Imaging

Combing the high contrast of optical imaging and the high spatial resolution of ultrasound imaging, photoacoustic imaging(PAI) is a new modality for biomedical applications, such as breast tumor detection, functional brain imaging and blood oxygenation mapping, etc. With the demand of different clinical applications for the imaging of organs or tissues, three submodalities of the PAI system have been studied recently including photoacoustic tomography(PAT), photoacoustic microscopy(PAM) and photoacoustic endoscopy(PAE).The acoustic transducer and imaging reconstruction algorithm are key components of the PAI system. On the one hand, most of PAIs use the conventional acoustic transducer based on piezoelectricity. Although an acoustic sensor fabricated with conventional technologies could be used easily in a PAT system, it is hard to be applied for the other two PAI implantations. PAM can provide a lateral resolution ranging from a few hundred nanometers to a few micrometers. As such, a corresponding ultrasonic transducer working at a center frequency of at around 100 MHz would be required for each PAM system. Nevertheless, it would be expensive to fabricate such kind of ultrasonic transducer in order to obtain such a high center frequency for conventional technologies. For the PAE, in order to shorten time for collecting imaging data and realize real-time imaging for operational applications, an acoustic sensor array must be accurately arranged together in the endoscopy probe, which is also difficult for conventional acoustic transducer. On the other hand, although many reconstruction algorithms have been proposed, most of these algorithms based on a hypothesis that the acoustic speed of tissue is homogeneous. But this hypothesis does not consistent with the actual situation, because the acoustic speed of different tissue is different. Moreover, these algorithms are restricted to specific scanning mode, which is unsuitable for practical application. This thesis focuses on above mentioned problems and the main work is the research of ultrasonic transducer and image reconstruction algorithm in PAI. The main research works completed in this thesis include:1. A computer simulation platform is set up for the study of piezoelectric micromachined ultrasonic transducer(PMUT). Firstly, the performance of three PMUTs with different structure was simulated and compared, and then the physical parameters of three structures were analyzed in detail. Finally, a photoacoustic/ultrasonic dual mode endoscopy based on PMUT was designed. The preliminary results obtained from different structures mignt provide a useful guidance for PMUT design.2. A time reverse image algorithm based on finite difference time domain(FDTD) was proposed for PAT, the focusing characteristic of time reverse was used to calculate the sound source distribution. This method is not restricted by scanning way and can be used in acoustic speed heterogeneity and homogeneous media, which is hard for the conventional image reconstruction algorithm.3. The signal preprocessing and image postprocessing were also researched to enhance the image quality in this thesis. Two low-pass filter and the wavelet transform were used to eliminate high frequency noise. The results show that three methods all can eliminate high frequency noise very well, and the proformance of wavelet transform is better than the other two low-pass filter, because the wavelet transform maintain the amplitude of signal and impove signal-to-noise ratio(SNR). After image reconstruction, singular value decomposition(SVD) and two-dimensional fourier transform were utilized to remove artifact. The resuls show that the two ways can eliminate artifact in a certain extent and improve the image quality.4. Finally, PAT was utilized to obtain image of ablated cardiac tissue and study the performance of damage evaluation of radiofrequency ablation for verifying algorithm. We did a check experiment with non-ablated cardiac tissue and ablated cardiac tissue using photoacoustic tomography system. The results show that the signal strength of ablated tissue is 3-4 times higher than the signal strength of non-ablated tissue, and the photoacoustic image can clearly show the boundary of the ablated tissue. They suggest that PAI has the capabilities of radiofrequency ablation detection and evaluation of myocardial injured size.The research work of this thesis provides a theoretical guidance for PMUT design, and expands a new method of photoacoustic image reconstruction in heterogeneity media. The exploratory study of ablated cardiac tissue imaging using PAT and the design of photoacoustic/ultrasonic dual mode endoscopy provide a new research approach for cicatricial tissue detection and ablation myocardium damage assessment.