Comparison of deep learning-based emission-only attenuation correction methods for positron emission tomography.

Journal: European journal of nuclear medicine and molecular imaging
Published Date:

Abstract

PURPOSE: This study aims to compare two approaches using only emission PET data and a convolution neural network (CNN) to correct the attenuation (μ) of the annihilation photons in PET.

Authors

  • Donghwi Hwang
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea.
  • Seung Kwan Kang
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea.
  • Kyeong Yun Kim
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea.
  • Hongyoon Choi
    Cheonan Public Health Center, 234-1 Buldang-Dong, Seobuk-Gu, Cheonan, Republic of Korea.
  • Jae Sung Lee
    Department of Biomedical Sciences, Seoul National University, Seoul, Korea jaes@snu.ac.kr.