Scholarlink

In continuous-wave electron paramagnetic resonance (CW-EPR) imaging, many spectral projections are required to accurately reconstruct four-dimensional (4D) spectral-spatial data, which are used for oxygen mapping in tumors. The ability to obtain fewer spectral projections with less degradation of the resultant image quality is desirable because acquiring many spectral projections leads to a longer time for EPR imaging acquisition. The algebraic reconstruction technique (ART) is one of the most powerful image-reconstruction techniques for 4D spectral-spatial imaging. ART can be applied to incomplete sets of spectral projections. However, data sets of many spectral projections can provide reconstructed images of higher quality. In this study, we aimed to enhance acquisition speed by randomly collecting spectral projections and subsequently synthesizing those that had not been recorded. We applied this strategy to a numerical phantom and a mouse Hs766T xenograft model to confirm the feasibility of our concept. Random acquisition can prevent image degradation in linewidth mapping, which is the foundation for oxygen mapping with CW-EPR.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Random Acquisition of Spectral Projections in Four-Dimensional Spectral-Spatial EPR Imaging: A Feasibility Study Toward Oxygen Imaging

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Misa Oba,
  • Mai Taguchi,
  • Shingo Matsumoto,
  • Hiroshi Hirata

摘要

In continuous-wave electron paramagnetic resonance (CW-EPR) imaging, many spectral projections are required to accurately reconstruct four-dimensional (4D) spectral-spatial data, which are used for oxygen mapping in tumors. The ability to obtain fewer spectral projections with less degradation of the resultant image quality is desirable because acquiring many spectral projections leads to a longer time for EPR imaging acquisition. The algebraic reconstruction technique (ART) is one of the most powerful image-reconstruction techniques for 4D spectral-spatial imaging. ART can be applied to incomplete sets of spectral projections. However, data sets of many spectral projections can provide reconstructed images of higher quality. In this study, we aimed to enhance acquisition speed by randomly collecting spectral projections and subsequently synthesizing those that had not been recorded. We applied this strategy to a numerical phantom and a mouse Hs766T xenograft model to confirm the feasibility of our concept. Random acquisition can prevent image degradation in linewidth mapping, which is the foundation for oxygen mapping with CW-EPR.