Your browser doesn't support javascript.
loading
Synthetic CT generation for pelvic cases based on deep learning in multi-center datasets.
Li, Xianan; Jia, Lecheng; Lin, Fengyu; Chai, Fan; Liu, Tao; Zhang, Wei; Wei, Ziquan; Xiong, Weiqi; Li, Hua; Zhang, Min; Wang, Yi.
Affiliation
  • Li X; Peking University People's Hospital, Beijing, China.
  • Jia L; Shenzhen United Imaging Research Institute of Innovative Medical Equipment, Shenzhen, China. lecheng.jia@cri-united-imaging.com.
  • Lin F; Zhejiang Engineering Research Center for Innovation and Application of Intelligent Radiotherapy Technology, Wenzhou, China. lecheng.jia@cri-united-imaging.com.
  • Chai F; Shenzhen United Imaging Research Institute of Innovative Medical Equipment, Shenzhen, China.
  • Liu T; Peking University People's Hospital, Beijing, China.
  • Zhang W; Peking University People's Hospital, Beijing, China.
  • Wei Z; Shanghai United Imaging Healthcare Co., Ltd. Shanghai, Shanghai, China.
  • Xiong W; Shenzhen United Imaging Research Institute of Innovative Medical Equipment, Shenzhen, China.
  • Li H; Shanghai United Imaging Healthcare Co., Ltd. Shanghai, Shanghai, China.
  • Zhang M; Shenzhen United Imaging Research Institute of Innovative Medical Equipment, Shenzhen, China.
  • Wang Y; Peking University People's Hospital, Beijing, China.
Radiat Oncol ; 19(1): 89, 2024 Jul 09.
Article in En | MEDLINE | ID: mdl-38982452
ABSTRACT
BACKGROUND AND

PURPOSE:

To investigate the feasibility of synthesizing computed tomography (CT) images from magnetic resonance (MR) images in multi-center datasets using generative adversarial networks (GANs) for rectal cancer MR-only radiotherapy. MATERIALS AND

METHODS:

Conventional T2-weighted MR and CT images were acquired from 90 rectal cancer patients at Peking University People's Hospital and 19 patients in public datasets. This study proposed a new model combining contrastive learning loss and consistency regularization loss to enhance the generalization of model for multi-center pelvic MRI-to-CT synthesis. The CT-to-sCT image similarity was evaluated by computing the mean absolute error (MAE), peak signal-to-noise ratio (SNRpeak), structural similarity index (SSIM) and Generalization Performance (GP). The dosimetric accuracy of synthetic CT was verified against CT-based dose distributions for the photon plan. Relative dose differences in the planning target volume and organs at risk were computed.

RESULTS:

Our model presented excellent generalization with a GP of 0.911 on unseen datasets and outperformed the plain CycleGAN, where MAE decreased from 47.129 to 42.344, SNRpeak improved from 25.167 to 26.979, SSIM increased from 0.978 to 0.992. The dosimetric analysis demonstrated that most of the relative differences in dose and volume histogram (DVH) indicators between synthetic CT and real CT were less than 1%.

CONCLUSION:

The proposed model can generate accurate synthetic CT in multi-center datasets from T2w-MR images. Most dosimetric differences were within clinically acceptable criteria for photon radiotherapy, demonstrating the feasibility of an MRI-only workflow for patients with rectal cancer.
Subject(s)
Key words

Full text: 1 Collection: 01-internacional Database: MEDLINE Main subject: Rectal Neoplasms / Radiotherapy Planning, Computer-Assisted / Magnetic Resonance Imaging / Tomography, X-Ray Computed / Deep Learning Limits: Adult / Aged / Female / Humans / Male / Middle aged Language: En Journal: Radiat Oncol Journal subject: NEOPLASIAS / RADIOTERAPIA Year: 2024 Document type: Article Affiliation country:

Full text: 1 Collection: 01-internacional Database: MEDLINE Main subject: Rectal Neoplasms / Radiotherapy Planning, Computer-Assisted / Magnetic Resonance Imaging / Tomography, X-Ray Computed / Deep Learning Limits: Adult / Aged / Female / Humans / Male / Middle aged Language: En Journal: Radiat Oncol Journal subject: NEOPLASIAS / RADIOTERAPIA Year: 2024 Document type: Article Affiliation country: