Adaptation of image reconstruction algorithm for purposes of ultrasound transmission tomography (UTT)

Abstract

In this research, the convolution and backprojection method has been adapted for the purposes of image reconstruction in ultrasound transmission tomography (UTT). In particular, a complete computer algorithm enabling the use of different convolving and interpolation functions has been developed. A technique of reducing and scaling the tomographic measuring data to minimize the reconstruction errors is proposed. The convolution and backprojection method was optimized through a choice of a versatile convolving function and a simple interpolation function, and it was tested using simulated and actual tomographic measuring data. After reconstruction, good-quality images were obtained. It has been found that the number of measuring rays determines the resolution of an image, the accuracy with which the size of structures is imaged and the accuracy with which the image point values are reconstructed, whereas the number of measuring projections determines the dynamics and distortion of an image. Because of computation time and image blur, an optimum reconstruction grid size, according to the visualization quality criterion, should be chosen. The convolution and backprojection algorithm optimized in this research can be applied directly to the UTT visualization of the internal structure of objects as a distribution of local sound velocities in this structure, reconstructed on the basis of measurements of mean times of the passage of an ultrasonic wave through a cross-section of an object immersed in water, in a parallel ray projection geometry.