AIP Advances (May 2021)

Ultrasound tomography reconstruction algorithm using two opposite rotatable linear arrays for breast cancer diagnosis

  • Yijun Cheng,
  • Chang Liu,
  • Chenyang Xue,
  • Binzhen Zhang

DOI
https://doi.org/10.1063/5.0051052
Journal volume & issue
Vol. 11, no. 5
pp. 055124 – 055124-9

Abstract

Read online

This article presents a new ultrasound transmission tomography reconstruction algorithm based on a isometric fan-shaped beam along the straight trajectory for an original breast ultrasound imaging system with the cylindrical motion of opposite linear arrays. The ultrasound computed tomography system and measurement scheme are described in detail. The reconstruction algorithm is derived. The numerical breast phantom model is built, including eight structures representing the adipose tissue, glandular tissue, cysts, fibroadenomas, cancer, and calcifications. Using one array as a transmitter and the other one as a receiver, the attenuation distribution in a medium interposed between them is reconstructed. However, limited angle tomography is usually plagued by artifacts along the depth of the image that cause shape distortion to be ellipses, which had an impact on the accurate judgment of the size and shape. B-mode ultrasound images are used as prior information to aid in improving the image quality of limited aperture reconstructions. Simulation results show that the number, location, size, and shape of cancer, cyst, fibroadenoma, and calcification in the numerical breast phantom are reconstructed. Moreover, the quantitative analysis of reconstructed images demonstrates that the reconstruction performance of the region of interest has been greatly improved, which verifies the feasibility of the algorithm using two opposite linear arrays. The reconstruction images are produced by the designed ultrasound tomography reconstruction algorithm, allowing the assessment of the morphology, orientation, internal structure, and margins of lesions in the breast tissue. The ability to create good attenuation images using the cylindrical motion of opposite linear arrays allows reconstructing the three-dimensional image for breast cancer detection and characterization studies.