Transfer learning-based synthetic ct generation for mr-only proton therapy planning in children with pelvic sarcomas

Chuang Wang, Jinsoo Uh, Xiuxiu He, Chia Ho Hua, Sahaja Acharya

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Proton therapy planning requires Hounsfield unit (HU) data from CT images to calculate dose and, in the case of pelvic sarcomas, accurate registration of the CT to MRI to delineate tumor. MR-only proton therapy planning would eliminate the uncertainty associated with CT/MR image registration and the need for CT, reducing exposure to radiation and anesthesia in children. We determined whether MR-only proton therapy planning is feasible by introducing a transfer learning-based cycleGAN (TL-cycleGAN) method to convert pelvic MRI to synthetic CT (sCT) for dose calculation and to specifically address the challenge of a small training dataset, commonly associated with pediatric studies. The TLcycleGAN was designed to transfer knowledge gained from converting a large number (n=125) of pediatric brain MRI studies to sCT and finetune the well-trained model on pelvic data. Sixteen patients (aged 1.1-21.3 years, 7 females) who received proton therapy to the pelvis were randomly divided into training (n=11) and testing (n=5) groups. sCT generated from T1- T2-weighted fat suppression MR images was compared to the real CT in terms of peak signal-tonoise ratio (PSNR), structural similarity (SSIM) index, mean error (ME), and mean absolute error (MAE) in HU. The mean ± standard deviation of PSNR, SSIM, ME and MAE were 30.6±3.0, 0.93±0.04 -3.4±10.2 HU, and 52.4±17.6 HU, respectively, for T1W MRI; 29.2±1.5, 0.93±0.02, -6.6±24.8 HU, and 85.4±18.8 HU, respectively, for T2W MRI. Transfer learning facilitates MR-only pediatric pelvic proton therapy planning by generating highly accurate sCT for a small training dataset and a large variation.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2021
Subtitle of host publicationPhysics of Medical Imaging
EditorsHilde Bosmans, Wei Zhao, Lifeng Yu
PublisherSPIE
ISBN (Electronic)9781510640191
DOIs
StatePublished - 2021
Externally publishedYes
EventMedical Imaging 2021: Physics of Medical Imaging - Virtual, Online, United States
Duration: Feb 15 2021Feb 19 2021

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume11595
ISSN (Print)1605-7422

Conference

ConferenceMedical Imaging 2021: Physics of Medical Imaging
Country/TerritoryUnited States
CityVirtual, Online
Period2/15/212/19/21

Keywords

  • Cycle generative adversarial network
  • Deep learning
  • Pediatric patient
  • Proton therapy treatment
  • Synthetic ct
  • Transfer learning

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Atomic and Molecular Physics, and Optics
  • Biomaterials
  • Radiology Nuclear Medicine and imaging

Fingerprint

Dive into the research topics of 'Transfer learning-based synthetic ct generation for mr-only proton therapy planning in children with pelvic sarcomas'. Together they form a unique fingerprint.

Cite this