成像体模
材料科学
传感器
光学
滤波器(信号处理)
计算机科学
声学
物理
计算机视觉
作者
Lokesh Basavarajappa,Kenneth Hoyt
标识
DOI:10.1109/laus53676.2021.9639136
摘要
A new approach to photoacoustic (PA) imaging and tissue characterization is introduced. Termed PA graphic equalization (PAGE), this method divides recorded PA signals into a collection of frequency bands using a bandpass filterbank. Filter outputs are used to colorize a display. Lower frequency content depicted in the PAGE image are from smaller optical absorbers whereby higher frequency information is from larger objects. In this study, we performed simulations and experiments using homogeneous phantom materials. PA signals in radiofrequency (RF) format were simulated using MATLAB and the k-Wave toolbox (MathWorks Inc). Ultrasound (US) sensors were defined to mimic a linear array transducer. Spherical optical absorbers of diameter 8, 40, and 80 μm were defined as source elements. Simulated data was reconstructed using a time reversal method to obtain final images. Experimental PA data were obtained using a Vevo 3100 with LAZR-X system (FUJIFILM VisualSonics Inc) equipped with a MX250D linear array transducer. Experiments were performed using custom phantoms embedded with fluorescent microspheres with diameters of 10 to 45 μm or 106 to 125 μm. Simulation results from a numerical phantom containing different-sized optical absorbers of varying diameter revealed that PAGE images give insight into object size and information not provided by traditional PA images. PAGE imaging of phantoms with varying-sized optical absorbers found a 2-fold difference in mean PAGE image intensity between the two materials (p < 0.001). Conversely, PA images from these same phantoms did not exhibit any differences in intensity (p = 0.82). Overall, simulation and experimental results verified PAGE imaging can differentiate micrometer-sized optical absorbing objects of varying size.
科研通智能强力驱动
Strongly Powered by AbleSci AI