Research On Signal Processing In Acoustic Imaging Optimization

Medical imaging,as an important means to assist doctors in clinical diagnosis,has become one of the most widely studied fields due to its importance and universality.Researchers in this field have been expanding their research to high-quality,non-invasive and low-cost.In recent decades,medical imaging has developed from the original X-ray imaging,to the current magnetic resonance imaging(MRI),computed tomography(CT),ultrasound imaging(US)and photoacoustic imaging(PA).Among them,ultrasonic imaging and photoacoustic imaging have become a research hotspot in the field of medical imaging due to their non-invasive,low-cost,rapid and portable characteristics.However,limited by the propagation characteristics of sound waves,the shortcomings of imaging algorithms and the development of hardware,acoustic imaging has some problems such as small imaging depth,a large number of artifacts and low resolution.Image quality is of great significance for clinical diagnosis and treatment.Therefore,this paper proposes an image optimization method to solve the problem of target aliasing and low resolution in acoustic images.Considering that the photoacoustic signal and ultrasonic signal are essentially the superposition result of signals at multiple particles,it can be regarded as the convolution result of the signal at a single particle and the positive signal which representing the size of target.Therefore,the method proposed in this paper is based on the deconvolution algorithm,followed by the empirical mode decomposition algorithm,to process the original signals which detected by the ultrasonic probe,and then reconstruct the processed signals into images according to reconstruction algorithm.Among them,the convolution kernel used in the deconvolution algorithm is obtained by detecting the signal of the object whose diameter is close to the minimum size that can be measured by the ultrasonic probe.The method proposed in this paper can reduce the signal alising,and finally reduce the aliasing in the image,so that the originally unrecognizable tiny structure and detailed information in the image can be seen.The method proposed in this paper is effective to both photoacoustic signal and ultrasonic signal.To prove the effectiveness,this paper carried out simulations and experiments respectively for photoacoustic imaging and ultrasonic imaging.The simulation part was conducted on k-wave toolbox.The experiment part use Verasonics acquisition platform for data acquisition and partial imaging processing,and Opotek was used as the laser source in the acquisition of photoacoustic signals.The image optimization effect is presented by contrast.On the premise of using the same imaging algorithm,the image quality of the original signal and the processed signal is compared.Through the comparison of multiple groups of simulation and experiment,the image quality reconstructed by the processed signal is better than that reconstructed by the original signal visually and from objective quantitative evaluation index,which both verify the validity of the method proposed in this thesis.