PMID- 37489854
OWN - NLM
STAT- MEDLINE
DCOM- 20230807
LR  - 20230823
IS  - 2057-1976 (Electronic)
IS  - 2057-1976 (Linking)
VI  - 9
IP  - 5
DP  - 2023 Aug 4
TI  - Generation of synthetic CT from CBCT using deep learning approaches for head and 
      neck cancer patients.
LID - 10.1088/2057-1976/acea27 [doi]
AB  - Purpose.To create a synthetic CT (sCT) from daily CBCT using either deep residual 
      U-Net (DRUnet), or conditional generative adversarial network (cGAN) for adaptive 
      radiotherapy planning (ART).Methods.First fraction CBCT and planning CT (pCT) 
      were collected from 93 Head and Neck patients who underwent external beam 
      radiotherapy. The dataset was divided into training, validation, and test sets of 
      58, 10 and 25 patients respectively. Three methods were used to generate sCT, 1. 
      Nonlocal means patch based method was modified to include multiscale patches 
      defining the multiscale patch based method (MPBM), 2. An encoder decoder 2D Unet 
      with imbricated deep residual units was implemented, 3. DRUnet was integrated to 
      the generator part of cGAN whereas a convolutional PatchGAN classifier was used 
      as the discriminator. The accuracy of sCT was evaluated geometrically using Mean 
      Absolute Error (MAE). Clinical Volumetric Modulated Arc Therapy (VMAT) plans were 
      copied from pCT to registered CBCT and sCT and dosimetric analysis was performed 
      by comparing Dose Volume Histogram (DVH) parameters of planning target volumes 
      (PTVs) and organs at risk (OARs). Furthermore, 3D Gamma analysis (2%/2mm, global) 
      between the dose on the sCT or CBCT and that on the pCT was performed.Results. 
      The average MAE calculated between pCT and CBCT was 180.82 +/- 27.37HU. Overall, 
      all approaches significantly reduced the uncertainties in CBCT. Deep learning 
      approaches outperformed patch-based methods with MAE = 67.88 +/- 8.39HU (DRUnet) 
      and MAE = 72.52 +/- 8.43HU (cGAN) compared to MAE = 90.69 +/- 14.3HU (MPBM). The 
      percentages of DVH metric deviations were below 0.55% for PTVs and 1.17% for OARs 
      using DRUnet. The average Gamma pass rate was 99.45 +/- 1.86% for sCT generated 
      using DRUnet.Conclusion.DL approaches outperformed MPBM. Specifically, DRUnet 
      could be used for the generation of sCT with accurate intensities and realistic 
      description of patient anatomy. This could be beneficial for CBCT based ART.
CI  - (c) 2023 IOP Publishing Ltd.
FAU - Aouadi, Souha
AU  - Aouadi S
AUID- ORCID: 0000-0002-9839-2908
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Yoganathan, S A
AU  - Yoganathan SA
AUID- ORCID: 0000-0002-2857-8483
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Torfeh, Tarraf
AU  - Torfeh T
AUID- ORCID: 0000-0002-8758-9481
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Paloor, Satheesh
AU  - Paloor S
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Caparrotti, Palmira
AU  - Caparrotti P
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Hammoud, Rabih
AU  - Hammoud R
AUID- ORCID: 0000-0002-5537-3942
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
FAU - Al-Hammadi, Noora
AU  - Al-Hammadi N
AUID- ORCID: 0000-0001-5231-0017
AD  - Department of Radiation Oncology, National Center for Cancer Care and Research, 
      Hamad Medical Corporation, PO Box 3050 Doha, Qatar.
LA  - eng
PT  - Journal Article
DEP - 20230804
PL  - England
TA  - Biomed Phys Eng Express
JT  - Biomedical physics & engineering express
JID - 101675002
SB  - IM
MH  - Humans
MH  - *Spiral Cone-Beam Computed Tomography
MH  - *Deep Learning
MH  - Radiotherapy Dosage
MH  - Radiotherapy Planning, Computer-Assisted/methods
MH  - Image Processing, Computer-Assisted/methods
MH  - *Head and Neck Neoplasms/diagnostic imaging/radiotherapy
OTO - NOTNLM
OT  - CBCT
OT  - adaptive radiotherapy
OT  - deep learning
OT  - head and neck
OT  - patch method
OT  - synthetic CT
EDAT- 2023/07/25 13:08
MHDA- 2023/08/07 06:42
CRDT- 2023/07/25 08:43
PHST- 2023/04/16 00:00 [received]
PHST- 2023/07/25 00:00 [accepted]
PHST- 2023/08/07 06:42 [medline]
PHST- 2023/07/25 13:08 [pubmed]
PHST- 2023/07/25 08:43 [entrez]
AID - 10.1088/2057-1976/acea27 [doi]
PST - epublish
SO  - Biomed Phys Eng Express. 2023 Aug 4;9(5). doi: 10.1088/2057-1976/acea27.