Synthetic CT Generation Using MRI with Deep Learning: How Does the Selection of Input Images Affect the Resulting Synthetic CT?

Andrew P. Leynes; Peder E. Z. Larson

ICASSP 2018accepted0 citations

Synthetic CT Generation Using MRI with Deep Learning: How Does the Selection of Input Images Affect the Resulting Synthetic CT?

Abstract

Synthetic x-ray computed tomography (CT) images derived from magnetic resonance imaging (MRI) is a recent area of focus for medical imaging researchers for applications in attenuation correction in simultaneous PET/MRI systems and MRI-guided radiotherapy planning. Several research groups have demonstrated the potential of deep learning to generate the synthetic CT images, however, there are several major open questions that remain with this approach. We investigated how the selection of MRI inputs affect the resulting output using a fixed network. We found that Dixon MRI may be sufficient for quantitatively accurate synthetic CT images and ZTE MRI may provide additional information to capture bowel air distributions.

BibTeX

@inproceedings{icassp2018_syntheticctgener,
  title = {Synthetic CT Generation Using MRI with Deep Learning: How Does the Selection of Input Images Affect the Resulting Synthetic CT?},
  author = {Andrew P. Leynes and Peder E. Z. Larson},
  booktitle = {ICASSP 2018},
  year = {2018}
}